Show Notes
In this episode of Startups For The Rest Of Us, Rob and Mike talk about how to get your first 10 SaaS customers. They reference a ParseHub blog post about the beginning stages of their startup. Rob and Mike walk you through the steps of this article and how you might be able to replicate their success.
Items mentioned in this episode:
Transcript
Rob [00:00]: In this episode of ‘Startups for the Rest of Us,’ Mike and I talk about how to get your first 10 SaaS customers. This is ‘Startups for the Rest of Us’, episode 295.
Welcome to ‘Startups for the Rest of Us,’ the podcast that helps developers, designers and entrepreneurs be awesome at building, launching and growing software products. Whether you’ve built your first product or are just thinking about it. I’m Rob.
Mike [00:29]: And I’m Mike.
Rob [00:30]: And we’re here to share our experiences to help you avoid the same mistakes we’ve made. So, what’s the word this week, Sir?
Mike [00:34]: Do you want to hear a joke about tacos?
Rob [00:36]: Always.
Mike [00:37]: Are you sure?
Rob [00:37]: Uh-huh.
Mike [00:37]: It’s kind of corny.
Rob [00:39]: Well, you know. It comes with territory, right?
Mike [00:41]: Get it, corny? Tacos?
Rob [00:45]: I was waiting for something funny.
Mike [00:47]: I know. I don’t know what you were expecting from me.
Rob [00:49]: Ah, man. Yeah that was pretty bad. I wouldn’t even tell that one at MicroConf. That’s how bad that joke is. Yeah, we were talking offline before this podcast and it’s like we don’t have so much going – we have a lot going on, but none of it this week is that interesting to do in this update upfront. It’s like there’s behind the scene stuff. You said you’d talked to a potential customer about early access stuff. I’ve been plugging away having meetings, charging ahead. But I really couldn’t think of anything in the last week that was that interesting to tell the listeners about. Some weeks it’s like that.
Mike [01:21]: Yeah. It’s not that there’s not stuff going on. It’s just that it’s not really interesting to hear about at this point.
Rob [01:26]: Yeah, well so maybe today we just dive right in to the topic of the day. We’re talking about how to get your first 10 SaaS customers. And, specifically, we’re going to be referencing this blog post on the ParseHub blog called ‘How We Got Our First 10 Paying SaaS customers.’ And ParseHub is a SaaS app. They’re a visual web scraping tool. So, they have tag lines like ‘Extract Data from Any Site,’ or ‘Turn Any Dynamic Website into an API.’ And so, obviously, you sign up for them. You can go and it does screen scraping, in essence. And then you can pull data out. I’m assuming you hit some ParseHub endpoint and then you can pull data through there.
But they published a post – this is over a year old at this point. They had launched in September of 2014 and they published a post in 2015 about how they went through and – basically from a standing stop. As far as I know, they didn’t reference having a list, or they didn’t reference having an audience, or any of that stuff. And so that’s where I found it interesting – is that these folks have done it recently, relatively. They walk through – what is it, 5 or 6 steps that they took to do this. And we won’t cover in detail everything that they talk about. We’ll obviously link it up in the show notes. But we did want to kind of walk through some steps of how you might be able to kind of replicate what they’ve done to also get your first 10 SaaS customers.
And so the first thing they talk about is don’t rush. And the post mentions there’s a lot of pressure to launch early and grow from day one. But premature growth can be detrimental to a small team that’s not equipped to handle a large number of sign-ups. And, back to my voice here, not only can it be detrimental if you can’t handle the support and can’t handle the onboarding, but it can be detrimental if you have a product that nobody wants yet, and you launch and people try it, and then they don’t give you a second chance later on. Whether your product is just plain bad, if it’s not built well. Or if you haven’t built something people want yet it’s going to be really hard to get the benefit of the doubt and get another try out of folks six to twelve months down the line once you actually have built your killer tool.
And so ParseHub – in the posts – talk about how they didn’t launch publicly right away, that they focused on finding a pay-in point. They solved one problem for one type of customer very well. And this is where having this early access – I called it the “slow launch”. I mean, DRIP launched over the course of six months, in essence. We got customer number zero and then customer number one and we figured out “How do we built it until they get value?” And then customer number two. And we built up one at a time. And ParseHub, in essence, did a similar thing where they really took they’re time to focus on finding a real pay-in point and solving it for one target customer type.
Mike [03:53]: And I think in most cases that’s probably the way to go if you don’t have traction yet, and you’re not really sure if what you’re launching is really what people need. I mean, so there’s a difference between what your vision for the product is and what you’re able to deliver on day one. And there’s a lot of people who are going to be willing to put up with the early access bugs and the lack of features and functionality that you just aren’t going to be able to have on day one. And then there’s going be this large category of other people who are not willing to wait, and they don’t want to because they’ve got a problem right this second that they need solved and they need to have either you solve it or somebody else. And if you can’t solve it right then and there for them, they’re not going to be willing to wait.
So, I think that this point kind of drives home the idea that there’s this certain classification of customers, or prospects, that you have to have early on that are going to be willing to work through whatever those issues are over whatever that timeframe happens to be for you to get to a point where you can deliver not just everything that they need, but go above and beyond it, to the point that you’re able to attract and retain these other types of customers.
Rob [04:52]: Another reason that they’ve said don’t rush is that they’ve spent a ton of time up front solving this problem well. They said there are dozens of competitors that can extract data from simple, well-structured sites. But nobody could tackle the edge cases. ParseHub essentially spent a year – they said they grabbed a random sample of 30 scraping jobs from oDesk and they told themselves they wouldn’t launch publicly until they could handle at least 20 of them, and it took almost a year of development before they hit it. By that time, they’d built something really unique, really special. But what is they’re value proposition then? We can scrape sites that no other tool can. Like we are, quite literally, the best. We know that no one else can do that. So, they had a value prop that they’d developed through working with real customers over a longer period of time then most start-ups would be comfortable with. And – I haven’t done research into ParseHub – but I’m guessing by the fact that they took that much time that they were bootstrapping. Because if they had a big chunk of funding they probably would have been pushed to hire faster and get bigger faster. But instead they were really trying to build – in this case a better product – and seeking that product market fit in essence of building something people wanted. And it took them a long time. And it will very likely take that long when you’re doing it as well.
So, the second tip they offer is to perform extensive user testing with potential customers. They said they did more than 50 hours of user tests. And they said almost every change to their product was hinted at by what users did during an interview. They had friends, co-workers and friends of friends use the product while they watched. They found potential customers on marketplaces like Elance and oDesk and PeoplePerHour. They interviewed early adopters. They tested usability. They did all that stuff and they asked questions. Here’s a few sample questions they asked: “What web scraping techniques and tools did you try? What did you like or dislike about them? What kind of websites do you want to get data from? What do you need the data for? Is that data essential for your business?” And so, they said they focused on solving the customer’s problems. And they spent a ton of time doing it. So again, this is part of the don’t rush. It’s like, do it well. Don’t try to short-circuit this part of the process. We’re all in a hurry, and we’re all impatient and we want to get there, but realize that investing the time now will yield dividends if you really nail it in this early stage.
Mike [06:55]: And I think the point of what they’re trying to get at here is, specifically, that if you know what the customer wants before you try and build it then you can design towards the customer expectations, not what you think that they want, and then try to match it up later. At which point you’re going to have to go back and adjust for whatever that gap is between what you thought they wanted and what they actually were looking for. And that’s going to hold true regardless of whether or not you’re using the tool. And I think that if you are using like whatever the type of tool is that you’re building, if you’re in that target market, then you’re probably more prone to maybe make some assumptions about what the customer really wants. And that can be very difficult to adjust for the gap between where you ended up versus what they were really looking for, what they really needed.
Rob [07:39]: And the third tip they lend is to build something people will pay for. So, very much like building something people want. And I think the way they phrase it is they say, “Provide value that customers are willing to pay for.” In their post they say, “Finding a problem is great, but is it a problem that your customers are willing to part with their hard earned cash to solve?” They said before they had a product they would go on oDesk every day and apply for web scraping jobs. They asked people to commit to $150 a month for a tool they could use themselves. And once they got a few commitments, then they were sure there was a need to actually build software. And they point out that customers will pay you if you help them make money. We’ve said this many times on the podcast. Help them make money, help them save money, help them save time. And they found a product that, in essence, does all three.
Mike [08:24]: I think it’s very easy to kind of skip this particular piece of it and think that you’ve got this problem that needs to be solved, and it needs to be solved so people are going to be willing to pay for it. And it’s really a matter of whether or not they’re going to be willing to pay you for it. And the other side of it is whether you’re going to be able to find additional people who are also willing to pay for that. Back in late fall, when I started the process on Blue Tick, there was actually another idea that I was looking at and I really thought that that was the one that I was going to be able find traction with and get customers for, and it turned out that it wasn’t. It was an interesting problem. I think that there was a lot of money to be made there, but I don’t know if I would have been able to actually make that work, because I couldn’t find enough people that were – essentially a critical mass – that would have been able to develop into a customer channel. So I think that you have to have those conversations with people and find enough of them early on. And if you can’t go through the process and find even just 10 or 20 people who are willing to pay you for it, then what’s going to make you think that you’re going to get 100 or 500 or 1,000? So, if you try to think to yourself, “Oh, well, that’s a lot a work. I don’t want to do that.” Well, you’re not going to really be able to build a business if you don’t do that to begin with. So, you have to take a little bit of a step back and at least just take a baby step in that direction. Try and find the core group of users that you’re going to target. And if you can’t find them, you’re probably not going to be able to scale that into a viable business.
Rob [09:44]: They’re fourth tip is to leverage marketplaces and communities. And, with ParseHub, these guys focused on Elance and oDesk, but they specifically have a couple questions you should ask yourself. Number one: Where does my customer spend his or her time online? Two: Where does he or she look to find a solution for the problem I am trying to solve? And they give a couple of good examples. They say, “If you’re running a business that can help jewelry designers sell more product, reach out to them on Etsy?” If you want to solve a pain point in the real estate market, you can try kijijirealtor.com and other industry-specific communities. If you’re developing a solution for teachers, you can find them on Udemy or Skillshare.com. The idea here is that if you can find a community or a marketplace you’re not looking to scale it up to a bazillion users at this point. What you’re trying to do is get early feedback. And it gives you a little bit of exposure to people who are experiencing a pain point pretty dramatically, and if you’re solving that you are going to get some interest from folks. And that’s the point, you’re looking five, 10, 15 people at this point. Which the reason the title of this episode is ‘How to get your first 10 SaaS customers,’ so you’re doing stuff that doesn’t scale very well at all. But that’s the point. You’re just trying to hone in on their needs at this point so that you can solve them, and then you look to scale it later, and you look to replicate that out and expand it.
Mike [10:57]: One of the things that I really like, that they kind of eluded to in this particular articl,e which was they went out places like Elance and oDesk and looked specifically for job postings to solve this particular problem. They used those to help build their software, because it was evident from those job postings that people wanted that particular problem solved and they were willing to pay the money right this second for it. So, I think it’s worth going out there and taking a look at those different types of sites to find out if there are active job postings to solve the problem that it is that you’re trying to solve. That might also be a good place to just leverage to find ideas as well. I mean, if there’s enough people that are paying for a particular job over and over and over again, or particular problem they need solved, then you could, in theory, build software that is going to solve that problem for them. And – maybe you’re not going to be able to sell it to them because they need the solution right this second – but down the road if they need that problem solved again, or if there are other people who also come across that particular problem you can get them into a sales funnel and solve it later for them.
Rob [11:58]: Yeah, I agree. It’s a hack I don’t know that I’ve heard about it before, but it totally makes sense. These aren’t going to be easy things to solve. Imagine going onto oDesk and getting the list of all the people who want something extracted and then looking at those sites. I bet a bunch of them are really nasty, because I bet they had already tried all the solutions available on the market. And so you’re going to have to do some hard work here. You’re going to have to solve a hard problem. But that’s the nice part, is that you’re going to be able to charge for that if you are in fact able to solve it. I get the feeling so many of us as engineers, or of the engineering mind-set, or even just the entrepreneurial mind-set, once you have a problem you really want to tear into it. But it’s just finding that problem that people are willing to pay for. Even if it’s hard and you’re willing to spend six months of nights and weekends, or as these guys spent a year of customer development time. Finding the problem and vetting it is perhaps as challenging or more so than solving it itself.
The fifth tip from ParseHub is to engage on Q and A sites like Kora and Stack Overflow. And they talk about how they answer questions on Kora and Stack Overflow about web scraping and that this had a measureable impact on their revenues, especially in the early days. They also talk about launching on Hacker News Reddit and [Product Time?], which I think these days most people would probably try to do and they talk a little bit about. But I like the Kora approach which they reference – there’s a Kora how-to guide by Kissmetrics. And they talk about how there’s 40 underrated niche sites where you can post about your start-up and you can engage about that. This is – I’ll say it’s a somewhat common approach – it’s not completely novel or new. But if you can get on Kora or Stack Overflow and you do provide value and you can answer questions, because by this time you’re such an expert in your topic that probably off the top of your head you have more knowledge than 99+% of the people on the site. And so you actually can, off the top of your head, answer stuff with really in-depth numbers or thoughts or metrics because it’s just something you’re living and breathing. I have seen this with my products. Whether it’s DRIP today, or HitTail previously. I have seen getting customers from these types of things, because what’s interesting is that these aren’t just forums off in the corners of the internet. These are big sites with a lot of traffic, and the people who search for these questions are desperately seeking an answer. And so if you can provide a good answer to it – it doesn’t even need to be that the answer is “Oh, use my product.” The answer can often just be here’s some insight, here’s some metrics, by the way I learned this as the founder of this product. And it lends you that credibility. I like this approach.
Mike [14:20]: The other thing that nice about that is, as you said, these sites have lots and lots of traffic. And it’s not just the immediate traffic from the people who are asking the questions. It’s more the long-term traffic over the course of a year, or two years, or five years, because those answers that you left several months or years ago will continue to be searched and indexed, and people are going to look for those. Because the same questions on forums tend to come up over again. So, even on Kora you’ll find that the same types of questions keep getting asked and they keep getting answered. And even on Stack Overflow, to some extent, there’s a lot of times where there will be one canonical answer. And that’s really what the community there strives to try and do. But a lot of times there’s variations of a particular answer, and you can go in and you can answer them. And as Rob said, you just drop some notes, “Hey, I learned this as the founder of X.” And people will go through those and they’ll take a look at it and if they still need the problem solved, or if they still have questions, they’ll click through and they will end up at your site, and hopefully that will lend itself towards another marketing channel. But I do know people who’ve been using Kora to some degree of success to try and help push their sales funnel further.
Rob [15:26]: Yeah, and you might that that Kora doesn’t scale – and it’s not going to scale infinitely – but you see entire brands built just on answering Kora questions. Like with Jason Lemkin and [SaaStr?]. He says that’s how he built, pretty much, the core of the SaaStr brand. And he’s answered thousand’s, I’m assuming – I’m sure you’re going to look and see how many he’s answered – but thousands of Kora questions. And he’s become kind of the defacto guy to be talking about B2B SaaS, and how to grow that because of that. So, while you may not have the time to do that, if you’re able to answer questions up front and then later on you can use the approach that Steli does at close.io where he says they’ll often put out a blog post. And then he basically records a video and then they turn it into both a video, audio and a blog post. And then the kind of editor – the guy who’s helping him do the content production – reads the post and gets the ideas out of it and goes onto Kora and figures out are there questions that pieces of this post answer? And then will answer it, whether it’s posting it all on Kora, or linking back. I’m not exactly sure how they do it. But There can be value that is a somewhat scalable thing, because it’s not using Steli’s time anymore, and you’re not using someone else to continue offering a lot of value to the audience and answering questions that they want answers to. And, in effect, getting some of that value back by the handful of people who are going to click back and potentially check out your product.
The sixth and final tip from ParseHub about how to get your first 10 SaaS customers is to use Twitter. And, specifically, they say that Twitter works magic, and so do your first few loyal fans. They talk about how they had a few users give them positive shout-outs on Twitter. And that after their initial launch, someone noticed them and reached out to write about them. And, to quote, the author says he wrote an honest comparison between ParseHub and Komodo Labs and he included the pros and cons of the first version of ParseHub. I agree. I have this love-hate – mostly hate – relationship with Twitter and Facebook for that matter. You need to use your head when you’re doing this. You can’t be on Twitter all the time, in my opinion. You need to be running your business. But, he’s right that monitoring Twitter and engaging with folks, especially in the early days, will earn you fans. It can earn you press. I’d say it’s more important in this early stage, as you’re starting, to get momentum. And he indicates that the opinion of your first few users matters more than gold. Reach out to them. Ask for their feedback. Engage them in your story, because then they become advocates of yours. And if they like you and they like what you’re doing, they’re the ones who are going to help spread the word. And so, engaging with them on Twitter is one way of doing that. Email is another. The nice part about Twitter, of course, is that it’s public and if folks are asking you questions or high-fiving you essentially, virtually through Twitter that’s a nice way to engage and kind of I guess broaden your reach so to speak.
Mike [18:05]: That’s also a double-edged sword because if they’re experience was a negative one then, of course, it’s out there publicly for the entire world to see throughout all of human history at this point.
Rob [18:13]: But most people, like if you really are working with them and they have a negative experience, they tend to do it via email.
Mike [18:18]: Right.
Rob [18:19]: Especially in the early days when you’re really hands on. I would hope that you could handle that one on one and –
Mike [18:23]: Sure.
Rob [18:23]: – help them move along. But you’re right. It is a danger.
Mike [18:25]: Right. I mean, I won’t say I take issue with this so much as it really like – I don’t think that Twitter is a silver bullet for every single business. I think it’s great that it worked for them and it’s great that they were able to land an honest review between them and somebody else. But that’s certainly not going to happen to everybody. In some ways it was a lucky coincidence, but in other ways if you are really working hand-in-hand with a lot of customers over the course of a year to build your product, then that’s going to happen naturally on its own, regardless of whether it’s Twitter or some other place.
Rob [18:55]: Yeah. I would agree with that. That’s a good point to raise. There is absolutely some serendipity in this. I think there are ways to tip that serendipity in your favor by providing amazing support and hand-holding experience up front. I think, if possible, working with influencers – whether you’re handpicking them or whether you’re just lucking into them – that’s always a big one. Because then you want to know that email list or a big Twitter following, they tend to want to create content to send to that list and, if you really knock their socks off, it often gives people an excuse to talk about you. It’s not a sure fire thing, for sure. And you’re going to spend a lot of time that’s never going to yield anything. But at least ParseHub, in their early days, said that they used it to great effect, or at least to get them to that 10 customer mark. And I think if you did all of their other things and didn’t use Twitter, I think you’d be just fine. I think Twitter is more or an icing on the cake if you’re already there and you know your audience is there as well and that that is going to be a viable channel, then I think it’s something to think about. But I wouldn’t prioritize it as high as any of the other points we’ve spoken about.
Mike [19:53]: Right. But I think that’s more because it’s so dependent upon those other things. I think that your comment about icing on the cake – I think that that’s very true just because of the fact that if you do all of the other things then Twitter will work well, or Facebook, or showing off your stuff publicly is going to work well because you’ve already worked with customers a lot. You’ve seen exactly what they’re looking for and what they’re willing to pay for, and you’ve done all of those other things. And it eventually leads you down the road of being more public about your stuff, and being more public about your interactions and the success that your customers are having, which then leads to these other things and these other success marks like the unbiased reviews and things like that that you can then highlight. So it becomes something of a snowball effect at that point, whereas if you were to start out with Twitter, for example, it just wouldn’t work because there’s nothing to base it on. You don’t even know if you have a viable product yet, or you’re solving a real problem that people have. So I feel like this is not just the last on the list because it just happened to fall there, but because it naturally falls there. It has to, because all of the other things have to be in place in order for that to really work well for you.
Rob [20:57]: So, again, we’ll link up the full blog post in the show notes. And, hopefully you’ve enjoyed this look today into how to get your first 10 paying SaaS customers.
Mike [21:05]: I think that about wraps us up for the day. If you have a question for us, you can call it into our voicemail number at 1-888-801-9690. Or you can email it to us at questions@startupsfortherestofus.com. Our theme music is an excerpt from “We’re Outta Control” by MoOt used under Creative Comments. Subscribe to us in iTunes by search for Startups and visit startupsfortherestofus.com for a full transcript of each episode. Thanks for listening and we’ll see you next time.
Episode 294 | Back of the Envelope Business Model Test
Show Notes
In this episode of Startups For The Rest Of Us, Rob and Mike talk about back of the envelope business model test. This episode is loosely based on chapter 2 of the book, Scaling Lean: Mastering the Key Metrics for Startup Growth. Some of the points discussed include defining your minimum success criteria and converting revenue goals to customer acquisition.
Items mentioned in this episode:
Transcript
Mike [00:00]: In this episode of ‘Startups for the Rest of Us,’ Rob and I are going to be talking about back of the envelope business model tests for revenue. This is ‘Startups for the Rest of Us,’ episode 294.
Welcome to ‘Startups for the Rest of Us,’ the podcast helps developers, designers, and entrepreneurs be awesome at building, launching, and growing software products; whether you’ve built your first product or you’re just thinking about it. I’m Mike.
Rob [00:25]: And I’m Rob.
Mike [00:26]: And we’re here to share our experiences to help you avoid the same mistakes we’ve made. How you doing this week, Rob?
Rob [00:30]: I’m doing pretty good. I’m coming off a series of split tests that we’ve been running on the homepage of Drip. And actually Zack on my team took that over recently. And after I had run a couple split tests with – basically the normal result of a split test is that you’re not going to have an improvement. If you run ten split tests you’re going to get improvement on one or two that’s significant. And so, my split tests were just chugging along and just taking forever to run. And then the first one Zack runs, he made some more dramatic changes, he saw a 41 percent improvement in the click-throughs. And so now we’re taking it another step and we’re actually installing – something we should have done from the start – but we’re installing the pixels. So we actually know if it’s not just click-throughs but it’s actually leading to trial sign ups and that kind of stuff. But it was pretty cool to see that kind of jump because that’s definitely a notable percentage.
Mike [01:19]: Nice. Now that you’re doing that and when you go back, are you going to track everything through because obviously you have to have some sort of a benchmark, right?
Rob [01:25]: Yeah. We’re going to track it through. We didn’t talk today about whether we’re going to rerun the same test now that we have the revenue pixel in place, or if we’ll just use this as the new benchmark and start from there. But either way this stuffs always fun. I geek out on it because it’s like the engineer – this is engineering marketing. Nowadays it’s called growth hacking but this is what we’ve been doing for ten, 12 years, where it’s like applying the engineering mindset to marketing. Which isn’t something that was commonly done, or maybe it just wasn’t talked about very much, except for by direct response guys before a few years ago when kind of this growth hacking thing became popular.
Mike [02:03]: Cool. Well, I just recently kicked off my Twitter ads for Bluetick again. And I’m hoping that they aren’t completely messed up like they were the last time. I think that I talked about that a little bit on the podcast, where just before MicroConf I had put a bunch of new ads out there on Twitter and people were tweeting back to me and commenting to me like, “You’re doing it wrong”. And I was just like, “What ae you talking about?” And I didn’t really think to go back and look to see what it was. I just thought it was people trolling a little bit. I got enough of them that I went back and took a look at it and, of course, the Twitter lead cards were all screwed up and it was like people had to click on them and then click on them again. It was just messed up so I had to redo not the entire campaign, but at least different parts of it. Hopefully things will go well this time.
Rob [02:42]: Good luck with that. It’s always touch and go when you first start any type of ad campaign, especially if it’s not proven because you just have to monitor it really closely. Once you get to that point where you have something that’s working and you have history behind it and numbers behind it, it’s so much more – I don’t know, comforting. Or it’s just less nerve-wracking I guess when you start it up. Because you generally know the range that it’s going to fall into. But at the start, man, you can turn it on and be paying crazy click amounts or you can just get no impressions and not know why and stuffs always frustrating when you’re trying to figure it out.
Mike [03:11]: What happened before was I was getting lots and lots of impressions but very few click-through rates. So I wasn’t actually paying for very much, which on one hand that was nice, but on the other hand I just wasn’t seeing any sort of results that I was looking for and I just hadn’t had time to go look at it at the time. At least now I have a benchmark of what I should not be getting.
Rob [03:30]: For sure. Cool. So what are we talking about today?
Mike [03:32]: Today what we’re going to do is we’re going to go through back of the envelope business model tests. And this is loosely based on chapter two of a new book that just came out called Scaling Lean: Mastering the Key Metrics for Startup Growth. And this is by Ash Maurya. And he’s written a couple of other lean startup books or at least books that are in that particular realm or genre so to speak.
But what I wanted to do was go through chapter two specifically and take a look at what some of the different thoughts are from him about how to look at a business model and determine what the forward looking plateaus are going to look like. And you can use this either for an existing business or for a brand new business that you’re getting off the ground. Some of the calculations that he has in the book are especially relevant just because they greatly simplify what he calls ‘customer throughput,’ which is your customer acquisition. And it kind of measures that against your customer churn rate as well, on a yearly basis. So taking those two things into account, you can sort of look at where your business plateaus are going to be and figure out whether or not you’re going to be able to have enough of a customer acquisition channel or channels in place in order to be able to just maintain the business – whatever your revenue goals are for the business.
Rob [04:43]: And keep in mind that unless you’ve run a business before and you kind of have some loose rule of thumb numbers that you use, some of the stuff we talk about today is going to be less applicable if you have no business at this point. Because you’re just pulling numbers out of the sky and you’re going to find that you’re pretty far off. But if you’ve already started and you have at least a few thousand a month in revenue, you have some numbers and you know your kind of trial to paid, and you know how something should go, you can be much, much more accurate with these calculations because you just have a concept of where you are and where you need to get to. Realizing that the numbers will change but in much smaller increments than if you’re just kind of throwing darts at a dart board as you would if you really do have a business with no customers to date.
The other thing I wanted to mention is keep in mind with books like this that they are written for a broader startup audience. And so not everything – if you do go buy the book, which I’m guessing is pretty good – if you do go buy it you got to take some things with a grain of salt. And we’ll try to point some things out specifically with chapter two, here, that I think may apply more to bootstrappers or ways that these could be shifted to people who are self-funded rather than the examples of the ten million dollar ARR after three years. It’s like that’s just so irrelevant to our audience in particular. We’ll try to call that out as we go through.
Mike [05:55]: Let’s dive right in. And the first step of chapter two is to take a look at the business itself and define what you would consider to be the minimum success criteria. And as I said before, the business model that’s shown in here can be applicable to either an existing business or to a new business that you’re trying to get off the ground and you’re trying to figure out whether or not it’s going to be a viable business. Take a look about three years out and try to think about what the business looks like specifically in terms of revenue. Now those two different things are extremely important. The first one is that you’re looking no more than three years out. And the second one is that you’re looking specifically at a revenue dollar amount so that you can make some sort of calculations.
If you try and go out further than three years, you’re probably not going to be nearly as accurate. And in addition, if it’s a business you’re trying to get off the ground, three years, trying to look beyond that and plan beyond three years is not going to be helpful to you because the business may just not be there or you may have a really hard time getting the customers. So looking at those in a much shorter time period, essentially time boxing your problem so to speak, is going to be really helpful.
The other side of it is being able to take a look at a firm dollar amount. You can adjust that later on, but the idea here is that you want to have a dollar amount in terms of the yearly recurrent revenue that you want to shoot for, that you can base most of your other calculations on. And we’ll talk a little bit more about how you can play with the numbers to kind of reach that revenue target based on lifetime value, customer acquisition rate, how long those customers stick around, etcetera. But those are just the two basic things that you need to think about when you’re talking about the minimum success criteria.
Rob [07:24]: And I’ll admit, to even think about thinking three years out feels crazy to me. It feels very MBA to even be talking about 36 months out because so much is going to change after you launch. When I’m first starting a business I tend to look six months out and think what can we get there. And if we hit product market fit where can we be at 12 months. With that said, this exercise, by the time we get to the end of it, it gives you a cool formula that I’m impressed with. It’s a high level way of trying to find plateaus and figure out where the business is going to plateau based on lifetime value and based on how many customers you think you can pull in.
So there’s a give and a take here. I don’t love the idea of looking three years out because I just think that you’re going to be way off. Even with the business as consistent as Drip, if I looked three years out I’m going to be way off and I know the numbers like the back of my hand. Take that with a grain of salt and realize that if you are projecting out and you’re thinking I need 120k per year, and maybe you want that after the first 12 months and you want that to be you quitting your job. But then you want the business – you’re thinking you want it to go to maybe double in the next year, and then go to 360 of 500k, kind of in that range is ambitious but I’ll say it’s not impossible for a bootstrapper to get there. Those are the kind of numbers that you should probably be thinking about in terms of this exercise. Assuming that you’re not going to have funding at the start and then you’re not going to try and grow a seven, eight figure business super-fast in the style of Silicon Valley; that you’re going to build it like a real business and hire based on profit and that kind of stuff.
Mike [08:56]: The second step is to take that revenue goal that you came up with and convert it into what he calls customer throughput. And this is your customer acquisition rate over time. And there’s a number of different steps to doing this. And the first step of that is if you’re building a new product you have to come up with some sort pricing on it. And if you don’t know what your pricing is, the recommendation is to use some sort of value based pricing to estimate the base price. This is essentially pricing your solution on the value of what it provides, not on what it costs to build and deliver. So that’s a difference between the solution value to your customers versus the cost structure on the back end to you.
Something else that you might look at is using cost based pricing, which is essentially taking your costs to deliver the solution, and then adding a margin on top of that. There’s a lot of business models that fit this particular mold of a service based model of any kind; productized services. Those tend to fit that. But those tend to fit that particular type of model. But if you can get away with it, if you can provide some sort of a value based pricing, you’re much better off. And going back to what we said before, if you have an existing business in place, you already have your pricing. You can essentially just use that number.
Rob [10:02]: For SaaS, I’d say it goes without saying that you’re going to want to shoot for value based pricing. That’s just kind of the way it’s done. You look at the value you’re providing, figure out if there is any competitors that are doing similar things and you either price above them if you want to be premium, or you price similar to them if you just want to be a better product. And I think another example – you mentioned consulting and such of using cost based – another example of that is kind of metered pricing, like how Amazon EC2 does it. I’m sure that they just look at their costs and then add some kind of margin on it. So, I think for the purposes of bootstrappers and folks listening in this, value based is 99.9 percent going to be the direction you want to go.
Mike [10:39]: The second step is to calculate the total number of customers at the end of the time period that you want in order to identify what your active customer base needs to be in order to make your ends meet for the revenue target. So let’s say that your revenue target is $180,000 a year, if you’re charging $30 a month then you need 500 customers in order to be able to reach that revenue goal of $180,000. That’s the kind of calculation that you need to be able to do. And this is why it’s so important to come up with not just the time periods but also what you are selling your product for and what your average price point is going to be for your customers. Obviously, if you have different pricing tiers then you kind of have to guess a little bit. So if your pricing tiers are $50, $100, and then $200 a month, your average price might be something like $75 a month or $90 a month. It really depends on where in the pricing spectrum the largest number of your customers fall. And obviously it can go in the other direction, too. You might have an average price point of $175 even though you have a bunch of people on the $50 a month plan. So take those things into account, but you’re trying to get down to an average price point per customer, and a lifetime value.
Rob [11:48]: Lifetime value is very hard to calculate if you don’t have a product. I think we’ll come back to that point. If you really are spit balling this, you’re going to have to use some rules of thumb and you’ll be off by a factor of two or three. If you already have a product this is much each because then you should know this like the back of your hand.
Mike [12:03]: I think if you don’t have a product at all, then using benchmarks from other similar companies to get to an estimate or just using what their pricing models look like – again, going back to what Rob said about determining whether or not you want to be a premium priced product or commodity based product or something along those lines. Just use a conservative estimate if all else fails. If you’re really not sure, come up with some sort of a conservative estimate for most of these numbers.
Now that said, once you have the numbers for your lifetime value and for the yearly target that you’re trying to reach, the calculation that he offers up is to get your customer throughput. And to do that you would take your yearly revenue, divide it by the customer lifetime value. And this comes out to the number of customers per year that you need to add into your business in order to be able to maintain the business at that level, at that point in time. Now that’s not on day one. It’s not on the 12 months in. It’s at that multi-year mark that you came up with in the beginning. So the recommendation was three years, if you’re using three years. And for sake of an example let’s go through that. If you’re trying to get to $500,000 a year at year three and you have a $50 a month product with a two-year lifetime value, your lifetime value is very easy to calculate, it’s $1200 lifetime value. But your customer throughput is that $500,000 divided by your lifetime value. And that comes out to 417 new customers per year that you need to add.
Now again, this assumes that your business is at year three. And if you just look at the raw numbers of the customers who are paying you on a monthly basis, your business would need 833 active customers to get to 500k in yearly revenue. But if you also take a step back and you look at that lifetime value, you’re churning out 417 of these customers every year. Which means that at 500k a year you need to add 400 every single year in order to just maintain the business at that level. And this is really where those calculations start to come into play and you can start figuring out where your plateaus are if you’re going to hit them at that particular level.
Rob [14:00]: And when I first saw this calculation, which again is called customer throughput, and it’s your yearly revenue target. So, like Mike said, that 500k divided by your lifetime value. And when I saw that my first question was why are we dividing by lifetime value? Shouldn’t we be dividing by the annual revenue per customer? And as Mike and I batted this back and forth offline, and in fact this formula works, the one that he’s given works. And he’s doing some clever math and canceling some things out, but suffice to say we tweaked around with different lifetime values, different lifetimes and different monthly price points and in all of them the math works. So, what I like about this is it’s a high level thing. Don’t get me wrong, this is not something that you’re going to sit down on day one and it’s going to dictate everything about your business. But what I like about this is it’s pretty fast to calculate.
And based on when I’ve launched products, you have a general idea of what your price is going to be. You know it’s going to be maybe around 30 bucks, or around 50, or around 75. You know that your average revenue per year should kind of be that based on what you’re launching into. And then you can always take a guess at your lifetime. When in doubt go with 12 months. That’s kind of been my rule of thumb for people who are starting a new business. You’re going to start off way lower than that when you kick off because you’re not going to have product market fit, your customer lifetime’s going to be like four or five months. But as you improve it you’re eventually going to hit that one-year mark and move beyond it. So a one-year lifetime is reasonable, and a two year means you’re doing pretty well.
In certain spaces like, let’s say Web hosting, where people just don’t churn out nearly as much, you might have a four year LTV or even a five year LTV. And big enterprise software is also like that. Maybe a HubSpot or a Salesforce, those guys have these really long customer lifetime values. And with lower price point software, typically let’s say average revenue fees are 20 to 99 bucks a month. You’re just going to have higher churn, you always will. So you’re going to have between, let’s say a one and three-year customer lifetime. So it’s pretty easy to kind of run a couple different scenarios on this. If it’s 50 bucks a month and you’re doing one year, then it’s $600 lifetime. And if you’re doing three years then it’s $1800. And then you can pretty quickly get an idea of how many customers you’re going to need to bring in each year in order to replace the people that are leaving and to maintain that revenue level.
This is not a projection of where you’re going. That’s a whole separate conversation. To project where you’re going you want to sit down with an Excel spreadsheet and it’s a whole different set of numbers. But what this is telling you is where the business is going to plateau based on your customer acquisition. So if you see this number of 400 new customers per year, if you’re already in your business and you’re trying to grow this thing, it’s going to be pretty obvious to you whether or not you can bring in 417 new customers per year. Because you know your numbers. And you know your traffic sources. And you know your trial to paid. And you know how many trials you get based on unique visitors and you can pretty quickly see you’re either going to be above that or below it and where you’re going to plateau. That’s the fun part.
From here I would actually take this estimate – it is a higher level, more ballpark estimate – and I would dive into real numbers so to speak, of like your exact churn rate. Because I have a big Excel spreadsheet that I use to do this but anyone can put this together if you have your true trial to paid and your true visitor to trial and your true first 60-day churn and post 60-day churn. It’s just much more complicated though, and it’s going to take you a few hours to put together. And you’re going to see an exact projection. But the cool part is that this one that you can throw together in like five minutes is going to be within the ballpark. Close enough that it’s a nice first cut to give you an idea of where your business is going to plateau if you’re accurate enough with your churn and your lifetime value numbers. So this could be more useful when you’re first starting out if you do use those benchmarks of other existing businesses you might be competing against. If you can get any idea about their pricing and their churn and that kind of stuff this can give you an idea of how many customers you need to acquire right up front. And just give a sanity check on, “Boy, can I really bring in 4,000 customers a year if that’s what it takes to maintain that revenue level?” It just stands as a decent five-minute sanity check, I think.
Mike [17:47]: The other thing that I think that this is really helpful in showing you is that because you have that high level number of – whether it’s 400 or 4,000 new customers that you need to add per year – you can backtrack a little bit and say let me divide that by 12 and figure out how many new customers I need to add per month. And let’s say that if comes out to 100. If you’re only adding two customers a month or three customers a month right now, then you know that looking forward to that particular point in time that it’s probably going to be really challenging to find enough customer acquisition channels to get from two to a 100. So it does give you that ballpark sanity check that you may need in order to be able to determine whether or not this is a business that is going to take you to where you want to go. Or whether it is something that you should probably offload and go look for a different business or just try a completely different business to start with depending on whether or not it’s an existing business that you have or an idea that you’re trying out.
Let’s move on to the next step. Once you have this customer throughput number, then you can go back and take a look at revising some of your previous estimates. And the first one that you can obviously adjust is that high level revenue target. That’s probably the last one that you want to adjust but it’s the first one that shows up on the list because that’s the high level, big, hairy, audacious goal that you’re trying to reach. You can adjust that; you could up or down. Chances are probably good that based on your estimates it will most likely end up going down. But that is one option.
The other option is to take a look at the lifetime value and try and figure out whether or not there are ways to either increase the lifetime of the customer, which is going to raise your lifetime value. Or raise prices in such a way that it also raises the lifetime value. And those are essentially the two ways that you can adjust this number. It’s either adjust that revenue target or increase the lifetime value. And those are really your only two options available to you.
Rob [19:30]: And again, if you’re doing this on paper before you started a business it’s harder because you’re just guessing at the LTV. But if you are a year or even six months into a business, you’re going to have a reasonable idea of your LTV and probably some ideas about how to increase that, whether it’s by reducing churn or increasing prices.
Mike [19:47]: The one thing I do like about this particular piece of it though is that – even if you are at a pre-revenue stage and you’re trying to validate things – if you have to look at your lifetime value and ten X it or 20 X it, or raise prices by ten or 20X then chances are good it probably points to the fact that this may not be a viable business model at all for you. Obviously, there’s probably other costs and stuff that you’re going to take into account. But again, you want to be using conservative estimates to begin with. So if these numbers do not pan out on paper then they’re probably going to be significantly more difficult to make work in real life.
That kind of leads us to our takeaway. And that’s the first takeaway. If you can’t make this business model work on paper then you’re never going to be able to make it work in real life, barring some form of miracle in terms of doubling your LTV or quadrupling it. Because those things are going to be very difficult unless your initial estimates were way off. Which is possible, but you also have to take into account that when it comes to math like this, if you have a bunch of estimates – there’s various theorems out there that say if you have all these different estimates or a number of different data points – chances are really good that your final number, because it’s an average, it’s going to come out in an average range. It’s not going to be at one of the extremes.
Rob [20:59]: And to give you ballpark ideas about lifetime values, it ranges very broadly because if your churn is high and your price point’s low, you can pretty easily have lifetime values in the $100 range. Like if you’re charging, let’s say $9, $19, $29 a month, if those are your tiers, you’re probably going to have a lifetime value that’s between – assuming you don’t have just crazy churn – you’re going to be looking at around somewhere between 80 bucks and a 150 bucks because that’s just where lower prices apps tend to churn out more than higher priced apps. And getting something into that range, let’s say that the 150 to 250 range is harder to do than you might think.
Now with that said, if you’re able to build an app that businesses depend on and that they really are using as kind of a core piece of their business, you can pretty quickly jump that above $1000 to $2,000 is completely reasonable for a business or for a SaaS app that has a monthly fee. Maybe the tiers are 50, 100, 150 and even on up for enterprise. If you’re building something people are relying on and they’re sticking around for a couple years – two to three years – that quickly gives you a lifetime value in that $1000 to $3,000 range, let’s say.
And moving on up, of course, you get a Salesforce or a HubSpot of whatever, which have lifetime values of tens of thousands of dollars per customer. And some even into six figures. And that is because the price points are so high and because they lock them into annual contracts, and because their entire business if focused on it. And so that’s your real range.
But if you’re listening to this podcast and you’re just starting out, you’re probably going to want to think my LTV’s going to be between 50 and 100 bucks. 150 bucks once you know what you’re doing. But it’s going to start out really low. Unless I’d say if you’re a repeat entrepreneur and you kind of know more of what you’re doing, you’re going to be able to push it into the several hundred dollars and then potentially into the low thousands. These are kind of ballpark guesstimates based on all the apps that I’ve seen.
Mike [22:44]: Another key takeaway is that the calculation for the customer throughput is really heavily dependent upon the inputs and the outputs to the model. And those inputs and outputs help you to define what is and is not actionable. So when you’re taking a look at the lifetime value, for example, that’s one of the inputs into this. And you can take action on that. You can raise the lifetime value of the customer by either increasing prices or increasing the lifetime of the customer.
In terms of the outputs, the number of customers that you need to reach on a yearly basis, you do have influence over that, and it is dependent upon the types of marketing channels that you use, advertising, whether you’re doing some sort of affiliates or leveraging other people’s networks. A lot of those things you have some level of control over. But obviously the math itself has to work. If you can’t make those final numbers work for you based on the inputs and the output, then the business model itself is not going to work for you at that point in time.
Rob [23:36]: I think some other things to keep in mind are this type of calculation, it estimates the viability of a business. It doesn’t give you an exact answer but it does give you a ballpark sanity check on whether or not this thing’s going to fly. In addition, Ash points out that time-boxed goals are more concrete and thus better than kind of simple revenue goals of I want to get to 500,000. It’s like without it being time-boxed what does that mean? And that’s something that I’ve always liked. I look ahead, like I said, six to 12 months and have a spreadsheet that’s looking at that. Because without the timeframe, it has so much less meaning because you have no concept of how many customers you’re going to need and how low your churn’s going to need to be. And therefore, how many trials you’re going to need. And therefore, how much traffic you’re going to need. If you know one step to the next what the conversion rates are, you can just back calculate from a 12-month revenue goal, you can back calculate to exactly how many unique visitors you need per month in order to make that happen. And if you’re complex enough you can include things like churn and upsells and downsells, and upgrade revenue and downgrade revenue.
It gets complicated but the idea is that this calculation that we talked about is that first swipe at it to give you the sanity check, and then you can dig into it as you get numbers that are better and that are real once you’re into the business. And then you can start plugging those in and figuring out how to improve them and how close your original estimates really were.
Mike [24:56]: Now I don’t know if this is something that Ash covers in a different section in the book, but one thing I think that we should probably talk about is a little bit about the difference between something like this versus your growth targets and your growth goals and how to look at those. Because what this gives you is that back of the envelope test to say at such and such point in time what does the business have to look like, how many customers do I need to acquire and how many will be leaving at this particular time. But that can be heavily overshadowed by your growth or your current growth. So if you’re growing at a very fast clip right now, it can be very easy to be distracted and look kind of micro focused at the business itself right now; how it’s doing, how you’re acquiring customers, doing split testing on all these different things and onboarding customers, doing support. And not really think about this down the road because you’re so hyper focused on that three to six-month timeframe.
But if you don’t take a step back and look at something like this then you can easily run into a situation where your business essentially flip flops and you almost drive it into the ground because you’re not paying attention. You’re hiring ahead of the curve or ahead of the need because the business is growing so quickly and you don’t realize that 18 months, 36 months down the road you’re probably going to run into serious customer acquisition problems or business problems because your customer acquisition needs are going to become so high based on your lifetime values.
Rob [26:12]: That’s the key here. These are not growth targets we’re talking about. These are plateaus. This is a heads up about a potential plateau. And this is something you need to be looking ahead at constantly as a subscription business. We had Ruben Gamez from Bidsketch on 50 episodes ago, I guess, to talk about how to identify and overcome plateaus. And this is the biggest hurdle that I see new SaaS founders hitting is not looking ahead and projecting. Given my churn, given my average revenue per user where are we going to plateau and how do we get past that? And the answer can be add more trials into the funnel. Sometimes that’s what it is. Sometimes the answer is we know that our funnel has no optimization so it may be running a bunch of split tests because you should be, at that point when you’re projecting, that you should be at scale. And I don’t even mean big scale like venture capital scale, but even if you just have 10,000 uniques a month or something, you can start running some split tests or even just making improvements on conversions on the site.
So there are a bunch of different ways to do it, but the idea is that when you’re running this business you have to be thinking ahead and projecting when am I going to hit the next plateau? And that’s what this calculation is about rather than growth projections. That’s probably another episode entirely.
Mike [27:24]: Sure. And then once you’ve identified what those plateaus look like and where they are likely to occur, then you can backtrack to where you currently are, plug in your numbers into a growth model and say, “when do I think that I’m going to end up actually hitting that plateau?” Because the business model might say it’s two years out or three years out, but looking at where your business is right now, if you’re growth rate is much higher then you could very well hit it in 12 months. And you really want to be in a position where you are looking at how to adjust the lifetime value of your customers well in advance of that. As Rob said, if you’ve done all those optimizations and then there’s not really other ways to address that, then you can start looking at your lifetime values and say, “how can I keep customers on longer? Are there ways for me to raise my prices or offer additional services?” And what that will do is that will increase your lifetime value, which will essentially push out that plateau even further.
Rob [28:15]: That wraps us up for today. If you have a question for us call our voicemail number at 888-801-9690. Or email us at questions@startupsfortherestofus.com. Our theme music is an excerpt from ‘We’re Outta Control,’ by MoOt. It’s used under Creative Commons. Subscribe to us in iTunes by searching for ‘startups,’ and visit startupsfortherestofus.com for a full transcript of each episode. Thanks for listening and we’ll see you next time.
Episode 293 | The 6 Biggest Email Marketing Myths
Show Notes
In this episode of Startups For The Rest Of Us, Rob and Mike talk about the six biggest email marketing myths. The episode is put together around an unbounce.com blog post. Rob and Mike discuss whether they agree or disagree with the myths mentioned in this article.
Items mentioned in this episode:
Transcript
Rob [00:00]: In this episode of Startups for The Rest of Us, Mike and I discuss the six biggest email marketing myths. This is Startups for The Rest of Us Episode 293.
Welcome to Startups for The Rest of Us, the podcast that helps developers, designers, and entrepreneurs be awesome at building, launching, and growing cell phone products, whether you’ve built your first product or you’re just thinking about it. I’m Rob.
Mike [00:28]: And I’m Mike.
Rob [00:28]: And we’re here to share our experiences to help you avoid the same mistakes we’ve made. What’s the word this week, sir?
Mike [00:33]: I talked a little bit about it last week, but we’re finally putting the finishing touches and final testing on the huge code overhaul that we’ve been doing over the past several weeks. It’s been a long trying process at best, I’ll say, but it’s finally almost done. We’ll be able to go live with it in a couple of days. It’s prevented us from pushing new code live just because of the way that we did things. And we too had a couple of missteps along the way, but we didn’t push anything live to the production servers because we went through the testing process and it’s like, “Oh, this just doesn’t work,” or, “It’s broken in this way,” so we held off on some of that. But hopefully another 24-48 hours and that’ll be over with and then we can start moving ahead.
Rob [01:07]: Those are always super trying processes, man, trying to get all the code up to speed. Now why did you have such an overhaul this early? Typically, I’d imagine you get technical debt and then you’d wait a year and you really run into it, and then you come back and revamp. But why only you’re kind of still in early access at this point and a few months into development?
Mike [01:26]: So it had more to do with how things were interacting with the database, and the fact that it was just not easy to write unit tests. So we basically skipped a bunch of things early on in the development. And the things that we skipped made it really, really difficult to put automated tests in place. And we made the conscious decision to not do some of those tests, but we took a step back and there were bugs that were getting into production and things that were being broken. So, I took a look at it and we went through and tried to figure out, “Okay, well how can we write unit tests and make sure that this doesn’t happen again?” and it was actually not possible to write those unit tests because of the way that some of the code was structured. So, in an effort to get us to the point where we’re not incurring more technical debt every single time we had a feature, we did a code overhaul of some of the way that stuff accesses the databases and external resources and things like that, because we interact with the mailboxes and we’re heavily dependent upon the current time. So, because of those things you have to have an abstraction layer in the place that essentially mocks those things up and says, “Oh well, if you were to access this thing, this is what that would return.” But because we didn’t have any of those abstractions in place, you literally can’t test them.
So, there was just a lot of…it wasn’t really spaghetti code, it was just a lot of internal dependencies on things that we don’t really have control over and they were embedded in the code. So, just made it very difficult, like I said, bordering on impossible to test some of those things in an automated fashion. We bit the bullet now to go through that to get us to the point where we can do that. And it’s not to say that we’ll do it in every single case moving forward, but any time we come across a bug, I want to be able to be sure that we can write a test that will make it easier to filter those things out. So, when we run into things in production, we can get rid of them and test it locally and make sure that we’re not going to run into that again.
Rob [03:14]: Very cool. It’s always a bummer to have to do that upfront. It kills time but I think it’s probably the right choice if it’s allowing you to write unit test. Because I think I’ve said several times over the past couple of years that I will never build an app or work on an app again that doesn’t have extensive unit tests, because it has saved our bacon so many times with Drip. Obviously, it takes longer to build the feature. It’s incrementally longer especially in the early days, but over the long run, it will save you time. So, I think you’re making a good call there.
Mike [03:43]: Like I said, it wasn’t so much about writing the unit tests, it was having the ability to. So, when we do run into something, we need the ability to write those so that we can test it and say, “Okay, this is what’s happening in production.” And then we need to be able to replicate it locally and we couldn’t do that. So, there was one bug we run into, and I talked about it last week, where I literally could not replicate it locally because I couldn’t get everything set up and mocked up in a way that it was in production. So, I was just guessing at different things to say, “Okay, will this work? Will that work?” And I had to push several different variations of the code in order to ‘fix it’, so that was really the issue. It was just when we run into something; we need to be able to test it locally to make sure that we’re fixing it.
So how about you? What’s up this week?
Rob [04:28]: Things have been pretty good. They’re kind of busy and chaotic right now. With school ending, I’m sure you ran into having a lot of end-of-the-year presentations and awards ceremonies and plays and parties. And just so happens both of our kids have birthdays around this time. So, I’ve just felt like we’ve gone to a lot of performances and a lot of events recently. Which is fun to go to, but it does kind of fill up the time and mean we have less of our evenings and weekends free right now.
Mike [004:52]: My kids had a half-day of school yesterday, which you’d think with two weeks left that they would still have full days of school. But for whatever reason, they had a half-day of school yesterday. So, you have to deal with it.
Rob [05:01]: Right. And then we’ve had a transition. We had a friend of ours, a nanny in essence, watching one of our sons, and she transitioned into doing yoga instruction full-time. So, we’ve had to transition him on that as well as he’s now done with school as of last week. So, we’re really just rushing around, getting stuff done, in addition to some other chaos going on.
So, today we’re talking about the six biggest email marketing myths. And I put together the outline for this episode around an unbounced blog post. And it was published in mid-May, so it was a couple of weeks ago, and it’s called The Six Biggest Email Marketing Myths Debunked. I wanted to talk through it, because some of them I do think are pretty good points that we should discuss that people should or should not do. And then some of them I just don’t think are actually myths. They may be some light rules of thumb, but I don’t think they’re things that everyone proclaims are rules and the truth and going against them is a terrible thing. But that’s the fun part, is that we’re going to be able to look through this.
So, the article kicks off and it talks about how there’s over 205 billion emails sent and received through the Internet. And it talks about that one McKenzie study suggests that email marketing is 40 times more effective at acquiring new customers than Twitter and Facebook combined. And we’ve talked about that on this show here, right? I typically say it’s a 20:1; I didn’t think about 40:1. But this goes back to when I had an RSS feed on my blog with 20,000 people and I had an email list of about 1,000 and when I launched both of them, the email sold as much as the RSS feed did. Which was surprising to me, because I thought that a lot of people were reading the blog. But it turns out they were reading it amidst a bunch of other blogs, and so they didn’t really know whose blog was who, because in an RSS reader it kind of all blends together.
And then, of course, the whole thing of having 20,000 Twitter followers versus even 1,000 or 2,000 email addresses I would say still holds true. So, it’s neat to see that someone has actually studied that. That’s why so many people – like we had a question last week of a founder who is saying, “It seems like all the advice I see -” which I think is a bit of an exaggeration, but he said, “I think all the advice I see it starts with the presupposition that you have an email address. And I think the reason is because email addresses they just work so well. And so it’s basically everyone encouraging you to go start building your list.”
And so let’s dive into the first myth here. And again, it’s the myth according to this article. I think, Mike, you and I want to discuss whether or not we think these are myths, in fact. But the first one is that Tuesday is the best day to send marketing emails. The article says, you’ve almost certainly heard this one and from my rule of thumb, Mondays and Fridays are typically not very good days to send email because on Monday people are just getting back from the weekend and their inboxes are often filled with three days of email. And on Fridays, a lot of people wind up – they either take Fridays off, so then stuff would fall to Monday. Or people are just starting to check out for the weekend. And so Tuesday, Wednesday, Thursday, for me, have always been my rule of thumb. And I wouldn’t say these are hard and fast, but in general, that’s kind of what I lean towards.
Mike [07:55]: I think that for this you really need to figure out what your audience is like. Because I would suspect that there’s a huge difference between B2C versus B2B communication here. I think on the weekends, your B2Bcommunication is probably going to drop off a cliff. But depending on your audience in that B2B space, it might be appropriate to send them. So, if you’re sending emails to entrepreneurs or small businesses, your chances are probably decent of being heard above the crowd. Especially among entrepreneurs because they’re probably checking their email on the weekend or at least to some extent. Versus if you’re sending the emails to people who work at Oracle or Microsoft or any of the larger businesses, they’re probably not going to be checking their email on the weekends. I think that you have to take some of these data points with a little bit of a grain of salt and just understand that it is highly dependent upon who your target audience is. It’s not just that Tuesday is the best or worst day. It’s really relative to who it is that you’re communicating with.
Rob [08:51]: That’s a good point, and in fact I had a B2C eBook at one point. This is probably seven, eight years ago, and the emails on the weekends converted better. And I just guess because that’s what people were thinking about at the time. I never figured out why, I just knew that they did get higher click through rates on the weekends.
The other interesting thing or an interesting exception is when I still had our apprenticelinemanjobs.com, which was a job board, I would see big traffic spikes on Sunday. And so I started sending more emails on Sunday, and it definitely made a big difference. Because, you know, on Friday or even during the week, folks aren’t necessarily thinking about trying to find a new job, but when Sunday comes, they’re starting to dread the next day’s work, and they’re thinking about how they can get into another line of work. It was crazy to see the traffic spikes. The biggest day of the week every week was Sunday and it was like Sunday afternoon, Sunday evening. So, that’s good thing to keep in mind.
What the blog post here talks about is there’s a HubSpot science of email report and they looked at the impact of the day of the week had on email open rates – so it’s only open rates – and what they found is that it’s crazy how smaller lists versus larger lists, there was a different impact. So, for smaller lists, Tuesday was actually the worst day of the week to send. And it wasn’t until you get to the larger list which is more than 10,000, so it’s not that big. But then Tuesday was a reasonable day. But to be honest, it seems like the article concludes here that if you send on Tuesday, you’re probably getting lost in the noise of the other email marketers who are also using this best practice of sending on Tuesday.
So, that’s the problem with a lot of advice, is that once everyone starts taking it, it doesn’t work as well anymore. The same is true with any of the marketing approaches that you hear about or any of like a blueprint or a roadmap to starting a startup or doing the marketing tactic. It’s like once everybody is doing it, then it doesn’t stand out anymore. It’s not remarkable and so it just kind of blends in to the background and you have to find that next thing. And so, I could see this one – we see big spikes Monday and Tuesday at Drip with our sending, in terms of our customers wanting to send email. I could see maybe thinking about shifting to Wednesday morning, Thursday morning sends.
But number two is that you can only send a particular email once. And this blog post talks about how you wouldn’t want to send the same email to the same people. But they talk about the tactic that Noah Kagan of AppSumo and SumoMe, and it says it was actually taught to him by Neal Taparia from EasyBib. It’s a tactic that Noah popularized where you take the same email that you have already sent to a list, and you change the subject line and you resend it only to your non-opens, right? So it’s people who didn’t open the first one and you change the subject line. And we’ve actually found this to be so powerful that we implemented it in Drip as a feature. When you’re sending a broadcast, there’s a single checkbox you check. And if you checked to resend to unopened, and then you can just say how many days you want to wait and what’s the new subject line. And it’s just all built in and it happens automatically, and it’s linked back to the original. And we have consistently seen increases of more than – well, between let’s say 25 to 35%, so in the 30% range. And that is what Noah had mentioned too, is that they were getting greater than 30% increase in opens from this tactic.
Mike [11:58]: I was going to mention exactly that when you guys added that into Drip and made it easy to do. You look through if you send out a bunch of campaigns; you can look back through them and look at the stats for those people who have been resent a particular email. And it’s astonishing how many of those people who didn’t open it the first time will open it the second time and it’s the exact same email. The only thing that’s different is the subject. So it’s pretty amazing that this little hack works and gets such significant results out of it.
Rob [12:25]: And you might think that you’re going to get complaints, and we have never gotten complaints in our use of it, and as far as I know other customers haven’t as well. The only time that I got complaints, I think I sent to my Software by Rob list and I accidentally – this was before we had this. This made us put some code in place to prevent this, but I had scheduled a broadcast with a resend to unopen and then I unscheduled it to make a change, and then I sent it through again with resend unopen. So I resent it twice. And I’m pretty sure the second two I didn’t differ the subject line between the two, and so then people said, “Why did you resend this email to me?”
But aside from that, which again is now not possible with Drip, right. We eliminated that because I did it to my own list accidentally. But even that one, I only got – it was literally three or four people who said, “Hey, how come you sent this?” and didn’t seem that big of a deal. So this tactic, I’m a big fan of it and we’ve only seen positive results from it.
The third myth we’ll talk about is keep your marketing emails short. And this is one that I’m not so sure that I’ve heard this as like a hard and fast rule. I guess I’ve heard enough people talk about the exact opposite. Like Patrick McKenzie often talks about how he sends out these very long emails and that his audience likes it. And same thing, Joanna Wiebe is actually – I’ve heard her say this, but she’s quoted in this article about how it’s not about picking one length or one style out of a hat, because it’s going to depend so much on your visitors and your prospects. And I’ve often heard in the Internet marketing spaces – especially 10 years ago – the sales pitch thing was: the longer the better. The longer you make your sales page, the more you’re going to convert. And I have found the opposite to be true.
We have a long forms page on the home page of Drip right now as an example, and we’ve done split tests with shorter versions of it. And the shorter versions appear to be having more traction. And so we’re looking at making changes there, and this happened remember the Microprenuer website, the home page used to be a lot longer. That’s over at Micropreneur.com. And I ran several split tests and I had a long, a short, and a medium, and the medium one just cleaned house both times and the content was generally the same. Like the headlines weren’t different. It really was just the length and it was cutting some stuff out. So, I think the same applies to email here. I certainly don’t disagree with it. I just wonder if it is really a hardcore rule. I haven’t heard that many people talk about it, so I question if it’s really a myth.
Mike [14:34]: And I guess it goes back to what I had said before about knowing who your audience is and what it is that you’re offering them, and at what point in the sales process. There are certain times where a much longer email is going to be warranted, and then there are sometimes where you just want to send a quick one.
And I think I remember one time – and it’s kind of a little bit of an aside, but we made a mistake when we were talking about MicroConf in one of our emails that we’d sent out to the list, and we accidentally put in an unsubscribe link right up at the top and the subject was MicroConf and people saw it. And as soon as they saw the email, they just clicked on the first link that they saw because it said MicroConf. And had we obviously put that that the bottom or made it a longer email with stuff at the bottom, then they would have had to scroll a little bit. But I think that knowing what it is that you’re sending them and why it is that you’re sending them that email makes a huge difference in whether or not your emails should be short or should it be long. Are you trying to explain something to them? Are you trying to educate them about a particular topic? Are you trying to get them to buy something? If you’re trying to redirect them to another location, then chances are good you probably want a shorter email. But if it’s just an education email, then there’s no reason that I know of, to go with a short one. You couldn’t certainly embed all that information there.
The other thing that you have to think about, I think, is the reusability of that information. So, if you have a series of blog articles and you embed them directly into an email sequence, then that’s perfectly appropriated to do. But just keep in mind that it’s probably going to be a little bit unwise to be sending people who are in your email list links back to that blog post if the reason they ended up on your email list was because they read that blog post. So, just be a little bit aware of contextually where those people are in the sales funnel, and how they ended up on your list, and what the content is that you’re sending.
Rob [16:20]: You make a good point about really the purpose of the emails. Because well, what I’ve seen with our up front email mini-course – which is Why Marketing Automation is the Future of Email Marketing – we’ve seen really good read rates and comment rates and reply rates on that course. And I think one of the reasons is that the posts are very palatable, and you can read them – I say the posts. It was originally a blog post and we turned it into an email mini-course. The emails are short enough, but they’re super palatable. I do think like in this case, having shorter emails that aren’t impossible to read on mobile or just take forever to read – if Patrick McKenzie gets away with giving people 15 to 20 minutes to read because his stuff is – he’s got a warm audience who really likes him and they know he’s going to give really valuable stuff, when you’re first engaging with someone, they may not give you that much time. In fact, most of them are not. And so being able to pack some really powerful actionable stuff into a short email, I think, is a good way to get acquainted with someone. And then once you get deeper into your courses, I think having longer emails can work. But keep in mind that just trying to throw 5,000 words in that first email – a lot of people they aren’t going to be up for reading all that.
The fourth myth is to keep your subject line short. And in this blog post, they reference a return path blog post that talks about how a typical desktop inbox displays about 60 characters of a subject line, while mobile devices show just 25 to 30 characters. And so there’s been some general advice in the past few years of keeping your subject lines under 30 characters. But what the return path found is that the highest open rate – well, they call it read rate. But I mean I’m not exactly sure they would know that. It’s really more open rate – is between 61 and 70 characters. It’s 17% at that point.
Now, it’s a bit of an anomaly because they go in 10 character blocks, and like the one below it is 14 and then the three above it are 14. And so the 61 to 70 really is this funky anomaly that bumps up 3 percentage points and not exactly sure why that is, but what that’s indicating is potentially there’s some magic. You know, they did look at 2 million email subscribers from over 3,000 retail centers for the month of February. And so there’s a non-trivial sample size. We’re doing this with 10,000 people or something. I would say that this could be more of an anomaly, but it seems like they’ve actually done some reasonable research here to point out that you don’t necessarily have to be super short. I don’t think 61 to 70 is a magical number, but I do think this points out that maybe short email subject lines are not the end all be all that some folks recommend.
Mike [18:43]: I don’t know very much about this one way or the other. I think that you have to have a minimum number of characters just to convey what the email is about. That’s the purpose of a subject line. But you also don’t want to run a War in Peace book in the middle of the subject line either, because there are going to be things that are cut off. And if it’s cut off, then it’s interesting looking at the data and the graph here that they have. It says they grab the average read rate and then the messages with this subject line length. And there’s not really a drop-off in terms of the read rate after a certain point. But if the email client cuts it off at 60 characters, then it almost doesn’t matter whether you have 60 characters or 60 million in the subject line, because they’re not being read anyway. So, people are judging it based on those first 60 characters. It seems to me like the data itself is a little misleading because you really have to take those types of things into account.
Rob [19:36]: Our fifth myth is that unsubscribes are bad. And in essence, the writer of this post talks about how people who brag about not having many unsubscribes are incorrect. And that unsubscribes are good because they remove people from your list who are unlikely to buy from you. And I would say in general, I agree with that. If you are emailing on a reasonable schedule and giving good content and someone unsubscribes, then okay, they probably weren’t going to buy from you.
However, I have seen some people who literally email every day, seven days a week, and have let’s say an autoresponder sequence of 60 or 70 emails long. And maybe when people unsubscribe from that list, it’s not that they don’t want to buy from you, it’s that they don’t want to hear from you 70 days in a row.
I’m not saying that that’s a bad tactic by the way. I think for the approach that this person was using, it was actually a reasonable thing for the goal they were trying to achieve. But what I’m saying is that I think unsubscribes are fine. I think that if you’re getting half percent of your list or 1% of your list to unsubscribe within any given send, that’s not a red flag, in my opinion, because are just going to come in and out of your list. The two things to keep in mind are one, if you do get an unsubscribe rate higher than that, why is that? What have you been doing or what did you do in this email that is encouraging people to leave your list? And two, make sure that you’re building your list fast enough. That having a reasonable unsubscribe rate, an expected unsubscribe rate, is something that you can overcome because you’re gaining more people than you’re churning out, basically.
Mike [21:02]: I think there’s a big difference in viewpoint here in terms of looking at the fact that you have people unsubscribing versus having an unsubscribe rate that is sort of out of control. And you talked a little bit about it where the number of people that you’re adding to your list is not overcoming the number of people that are unsubscribing, then you probably have an issue there. You are sending out an email every single day and your email unsubscribe rates are pretty high but you’re still adding a number of people in that are going to overcome that, that’s not necessarily a good thing either.
I do agree with you that I think that having unsubscribes is not necessarily a bad thing. There’s going to be people who opt out, and you do want those people to opt out if they’re not a good fit. I’ve seen marketers out there who will tell you flat out, “Hey, if you’re not interested, click here,” and they will actively make efforts to prune their lists to some extent to get people off of their email lists. And there’s a number of advantages to that. One is that it helps them to not be inundated with data from people that are just not interested and are never going to buy. The last thing you want is an email list of 20,000 people where 15,000 of them just don’t want to buy from you and are never going to. And then you’ve only got this group of 5,000 people. Well, three-quarters of your list is crap at that point. So what happens then is you end up with a lot of misleading data about what you think people want versus what they actually want. And three-quarters of those people you should not be listening to. The problem is which three-quarters of them is it, and you don’t know.
So, taking steps at that point to get rid of those people at that point is a good idea, but it comes down a lot more to segmentation than anything else. So, unsubscribes are not necessarily bad, but you also have to take into account the context of where people are unsubscribing and why.
Rob [22:42]: And I take it a step further with pretty much all of my marketing lists, and we use the feature in Drip called List Pruning. And this actually will remove people who are not opening or clicking or engaging with your emails, because even if they haven’t unsubscribed, if it’s just going into their empty inbox, that’s not good. It’s not good for some of these could be honeypots, Yahoo, Hotmail, Gmail. A lot of these mail providers will look at – and this is all black box stuff. So it’s rumored that they look at if you’re sending domains or your IPs are sending a bunch of emails and none of them are getting opened or it’s a very low open rate, then they consider your list lower quality. And so, it can hurt your deliverability long term.
And so, we do list pruning pretty regularly, and this is why it’s so funny. Like our lists are smaller than they would be if we weren’t pruning, but our open rates are way, way higher as a result. Because we keep our deliverability up and we only keep engaged people on the list. We do some joint ventures and we’ll partner up on an integration or something and people will tell us they have a list of x-thousand and the first question is asking is, “What’s your average open rate on a broadcast?” Because that’s way, way more important, because we’ve literally had people say, “I have a 40,000-person list, and their open rate is like 10 or 11% on a broadcast. Whereas bottom end you want to be above 20, in my opinion, and you’re solid and have a really healthy list if you’re between 25 and 35 for the broader lists.
And it depends about the age and the size and a bunch of stuff. But when you first start out and someone first stops in, that very first email, you want to see especially if they’re opting in to get some information. I want to see like a 70% open rate on that first email and then 60% at the low end. I have seen some that are up around the 80% range. And then over time as someone receives 20, 30 emails from me, that’ll slowly tick down, right. And so if you get 10 or 20 emails into an auto responder campaign, you’ll see it drop down into the 50s and the 40s. And then when it gets in the 30s, it should typically level out in that range, assuming you have a decent quality list. But if you continue to drop and you drop into the 20s and the teens, you have an issue. And so, not even just unsubscribes here, but I think just getting rid of people from your list who aren’t engaging, it serves a lot of positive purposes.
And our sixth and final myth is that marketing emails should be branded and polished. And frankly, this is something that I’ve been reeling against for years, is the big glossy fix with email newsletters thing with the drag and drop builder that you build this big Microsoft FrontPage looking thing and then you send that out to people. I’ve just found, in my experience, that having really nice plain text looking emails that are responsive. And they’re not actually plain text because you need to have the image in there to see if people will open, and you want to be able to rewrite your lengths and stuff. So it’s an HTML email, but it looks like plain text. I’ve been doing that for 10 years and I think that is the way to go with few exceptions. I think ecommerce is one exception where the visual element really helps. I think if you’re a recipe website where the visuals of the food, that’s where you can have something formatted. But to do this multi-column newsletter-type thing if you’re just communicating with your audience and you want to build a relationship with them, or even if it’s your customers and you want to have a relationship with them, I really have not been a fan of that approach.
Mike [25:49]: Did you really just reference a product that was discontinued in 2003 to send emails?
Rob [25:54]: What did I say?
Mike [25:55]: FrontPage.
Rob [25:57]: I did, because you know why I was using it as an example, we all know that using Microsoft FrontPage, you just build crappy web pages. And that’s what I see coming out of these drag and drop email builders is unless you know what you’re doing, you’re going to build a crappy looking email. So yes, I did. I hope it put the image in your head of exactly what I was saying.
Mike [26:17]: Yes. No, it did, but it also points back to – I remember using FrontPage as well and I was like, “Wait a second. FrontPage, is that even around?” I didn’t think that it was, and I looked back and it was discontinued in 2003, so that’s kind of scary that we both remember that and both have used it in the past.
Rob [26:32]: Totally. Totally.
Mike [26:33]: I do generally agree with this. I’ve seen lots of emails that come in. Some of them are branded and some of them are not. And to me it means more that the content itself is decent and that the source that it’s coming from is somebody that I trust versus how good it looks or how polished it is. That stuff to me doesn’t matter nearly as much as it does that the source of the email itself is somebody that I actually want to hear from, because if I don’t want to hear from them, then I don’t care how polished the email is. I just don’t want to.
If you look at a lot of the marketing emails that come from the Fortune 500 companies, they’re very well branded. And it’s not to say that the branding doesn’t help them, because it certainly does – it makes them recognizable, but they’re in a position where that matters to them. Versus if you’re sending your own emails, the chances are good that having a lot of that additional branding probably has very little effect on your business and it’s more about are you optimizing for the right things in your emails.
One of the contention points that I have about this particular item here is that some of the data they give is about plain text emails versus having HTML emails that have gif embedded into them. And they claim that there’s a 37% decrease in opens if you’re adding images into them. And I question whether or not that is a valid data point, because how would you even measure the open rate if you’re not including some sort of a tracking pixel of some kind in there?
Rob [27:54]: Right. And you can’t include that in a plain text email. And that’s what I was saying earlier, that when I say plain text, I mean an HTML email that looks like plain text, so that it allows you to rewrite the links so you can track clicks, and it allows you to embed the image so you can track opens.
Mike [28:07]: I think what’s probably more important is trying to figure out whether or not having a lot of brand and polish in there makes a difference. But even something like that, you’re not going to know until you get a fairly massive email list and have the ability to do some split testing on it. And at that point, it’s not so much about the open rates as it is about what the conversion rates are and the action that you’re trying to get them to take inside of the email. So if you just want to get your emails open, that’s one thing, but if you want them to go into the email, read it, and then take some sort of an action after they get the email, that’s a completely different story.
Rob [28:41]: Wow, this HubSpot study is pretty intense. They say with statistical significance specifically that they tested it with such broad, or I guess a higher number of emails that it is true. They come to a lot of conclusions like it says people say they prefer HTML, so if you give them a choice they’ll choose that. But over and over in every AB test they ran, they preferred plain text emails. They said that the HMTL emails reduced open rates versus the plain text looking ones I’m assuming. They said HTML emails have reduced click through rates. And they don’t just mean HTML. I think they actually mean – they have these big design emails with a lot of – there’s a lot of flair in there basically. And images and formatting, and the fix with type stuff I was referencing earlier with my FrontPage comment.
This is why, to be honest, the one template built into Drip is an HTML template that looks like plain text, and you can of course embed an image in it or something, but we have not encouraged people to do the big fancy templates. Although some people do bring them in, and I’m sure if they’ve run tests, they might find out with your specific audience that they do like some particular branding. We do have people who have elegant – they’re nice well designed, elegant templates, and those, I think, can help keep your brand in people’s minds. And I think they’re tasteful, they work with your audience. But I think just going with big fancy HTML by default is a really bad choice.
Mike [30:00]: And looking through at the emails that they have in here, some of the ones that they have side-by-side, they’re not terribly long, and if your email has five or eight lines of content in it and two or three massive images in it, to me it seems like that’s a very different story than if you have a longer post or you are using those images to help convey some sort of a story. Because I think if you a thousand-word email or something like that with a bunch of images in there for illustrative purposes, I think that’s very different than if you have three or four lines of text that have these massive images in it. And the main content or main focus of that email is to show you an image. To me those emails don’t really resonate. I would rather see some content along with it. But again, it depends on what the purpose of those emails is.
Rob [30:45]: I agree. And as I said before, I think if you’re Pinterest or you’re recipe or food site or a real estate site, there are reasons where visuals really can light it up and they should be the focus of something. But I think in general, if people are consuming your content and they want to read your content, you should stick more towards plaint text looking and then have images as needed in the email to break up the text.
Mike [31:06]: So, I think the bottom line with most of this stuff is to whenever you see some sort of a best practice or a guide about, “This is the best day to do this,” or, “This is the best way to do that,” take it with a grain of salt and think about what it means for you and what it means for your audience. Because by the time something becomes a best practice, it is not going to stand out anymore. And that’s something that you have to keep in mind, and figure out whether or not it’s still relative for our audience.
Well, I think that about wraps us up for the day. If you have a question or comment for us, you can call it into our voicemail number at 1-888-801-9690 or you can email it to us at questions@startupsfortherestofus.com. Our theme music is an excerpt from We’re Outta Control by MoOt used under creative commons. Subscribe to us in iTunes by searching for Startups and visit StartupsForTheRestOfUs.com for full transcript of each episode. Thanks for listening and we’ll see you next time.
Episode 292 | Moving from One-time to Subscription Revenue, When Your Core Product Has Variable Costs, and More Listener Questions
Show Notes
In this episode of Startups For The Rest Of Us, Rob and Mike answer a number of listener questions on topics including moving from one-time to subscription revenue, when your core product has variable costs, and how to disrupt the current payment method for a market. Mike also gives some updates and recent challenges with BlueTick.
Items mentioned in this episode:
Transcript
Mike [00:00:00]: In this episode of “Startups for the Rest of Us,” Rob and I are going to be talking about moving from one-time to subscription revenue when your core product has variable costs, and [more, some?] listener questions. This is “Startups for the Rest of Us,” episode 292.
[Theme Music]
Mike [00:00:20]: Welcome to “Startups for the Rest of Us,” the podcast that helps developers, designers and entrepreneurs be awesome at building, launch and growing software products, whether you’ve built your first product or you’re just thinking about it. I’m Mike.
Rob [00:00:28]: And I’m Rob.
Mike [00:00:29]: We’re here to share our experiences to help you avoid the same mistakes we’ve made. What’s the word this week, Rob?
Rob [00:00:33]: Well, MicroConf Europe is less than two months away, so if you’re interested in joining Mike and I in Barcelona with 100, 120 of your favorite bootstrapped friends, go to microconfeurope.com. We have a “buy a ticket” link in the upper right. We’d love to see you there. We have speakers that include Mike and myself and Steli Efti and Peter Coppinger from teamwork.com, and we’re currently working to recruit more speakers to the conference. We’re pretty stoked about it.
Mike [00:01:00]: Yeah, I’m definitely looking forward to it, although the time is going by really, really fast. I looked up the other day, and somebody had sent me an email: “Hey, by the way, see you in a couple of months,” and I’m just like, [groans] “Oh.” [Laugh]
Rob [00:01:09]: Yeah, it came up on us quick, so we have to get our other speakers in line pretty fast.
Mike [00:01:15]: Yeah. I think one of the downsides is just the timeline for us to sell tickets for MicroConf was pushed back a bit because of all the tax implications around accepting that for the conference –
Rob [00:01:25]: Yeah.
Mike [00:01:25]: – which, unfortunately, has to be also pushed onto the people who attend. We didn’t change the ticket pricing at all, but the tickets are priced higher now because we have to charge [VAT?] on everything.
Rob [00:01:35]: But then they can get refunds of [VAT?] –
Mike [00:01:38]: That is true.
Rob [00:01:38]: – so, ultimately, it won’t be more expensive. Almost everyone across the board will be able to get refunds, although I’m not an accountant. Don’t consider my advice concrete here, but that’s my understanding based on the consultants. It’s so complicated. We have to work with [VAT?] consultants. It’s crazy. You know you’ve grown up when you are hiring consultants just to deal with taxes.
What have you been doing the last couple weeks?
Mike [00:01:59]: Well, I’ve been working on on-boarding people into Bluetick, and I’m at the point where I’ve on-boarded over half of the prepaid customers so far. That’s good to see, but the downside is I ran into a bit of a minor setback. We ended up hitting a bug that caused a pretty major performance issue on my server. It was about three days or so – three, full days – and it was really hard to replicate in a local test environment, so we basically just bypassed the issue for a little while and said, “This functionality isn’t going to be available for you guys for a couple of days, or a week, or whatever until we get this straightened out.” We’ve had to refactor just a ton of code in order to get to the point where we can replicate those types of things locally. It just sucks to have to go through that, but I’d rather do it with under 20 customers than 200-plus.
Rob [00:02:49]: Yeah, now’s the time to do it. It is so painful early on, because you just want to move fast, because you’re just trying to prove things out and get that early revenue. But if you don’t make some good decisions here and eliminate that technical debt, not make the fast decisions where you have something hanging out there that later on can sting you, it’ll be brutal when you get into the hundreds; because having a major outage at that point sucks.
Mike [00:03:12]: Right. And it wasn’t even just that there would’ve been an outage. There’re things that, because of this bug, it just didn’t work. Or, at least there were parts of it that didn’t work. The rest of the application was still working, but it was causing certain jobs to be reprocessed over and over again, and it just pegs the CPU because of that. So, unfortunately, they just were never going to finish, which sucks. Having to take that step back and essentially put all forward progress on hold for close to two weeks – it really sucks, to be honest.
Rob [00:03:43]: Yeah, I know the feeling. It seems like with any app of any complexity these days, you’re going to run into this a lot in the early days because the code is just trying to ramp up. As you scale and you get more customers, we’ve found that about every – I’m trying to think. It’s probably every four to six months with Drip that we hit the point where we need to stop all feature development for a couple of weeks and just focus on fixing performance of individual – whether it’s queries, or whether it’s pages, adding caching, even upgrading the entire database to a bigger box with more RAM and all that stuff. We’ve cycled back on that to maybe three times a year, because you outgrow stuff. We’re not building these basic crud apps [laugh] like we used to be able to and compete. The stuff’s too competitive now. I think of a project management tool, or even time tracking, or invoicing software or something, you obviously want a lot of UX. There’s a lot that goes into building the product; but on the back-end the performance implications of it are really small compared to something that is sending a lot of email, or doing a lot of analytics, tracking opens, tracking clicks. Both of our apps do that, and any type of – I can’t even imagine what apps, like Mixpanel and Kissmetrics have to do on the back-end because they are the next level. Now I understand why those types of businesses had to raise funding. You just couldn’t get enough boxes and enough people to scale that up without having a big outlay in advance, even with a tool like Amazon EC2 or Rackspace Cloud.
Mike [00:05:12]: Yeah, I remember talking to Heaton at one point a while back about Kissmetrics, and one of the first versions of that that came out. I may have these numbers mixed up, but I think that he said that for 20 customers they had 12 servers, and it was just because of all the processing that they did.
Rob [00:05:29]: Yeah, I remember that. Obviously, they got better at scaling as it went out, but I do remember in some interview someone had said just the basic infrastructure, the fixed cost of their infrastructure – and I don’t remember if it was Kissmetrics or Mixpanel – was more than a quarter million a year. It might even have been half million. It was somewhere in that range just to keep everything running, and that wasn’t even to scale up as they got really big. That’s the kind of thing you have to think about. It’s getting easier and easier these days to build really cool tools, but the performance implications of those and the complexity of them as you scale up – it’s a real thing. I know that some startups have entire teams just devoted to keeping that stuff moving fast, and it’s a team of both Dev ops folks, DBAs and developers who are in modifying code just to keep things performing. You can imagine something real-time, like Uber, and how hard that would be since it’s worldwide and there’re so many incoming datapoints at any given time. The order of magnitude of complexity – that’s got to be incredible.
Mike [00:06:26]: The other side of that is that you have to have tools or systems in place such that you can test it and put it under artificial stress, and that’s something that we just haven’t really paid a whole lot of attention to because we were trying to move fast and trying to get the app out the door as quick as possible. We were able to do that, but the cost of doing so was that we didn’t have good mechanisms in place for us to be able to run absolutely everything locally and to integrate a lot of testing into the system. So, a lot of it’s been done manually, or there’re certain places that are much better-tested than others. It’s the places where we don’t have a lot of tests that really bit us, so right now we’re working on refactoring a lot of that code. Like I said, it’s been a full week. It’ll probably be another week before we get things settled down to the point that we can actually go back and start working on more features.
Rob [00:07:15]: Well, congratulations on getting half your folks on-boarded. I feel like this bump in the road that you’re hitting – that’s just the other shoe dropping. It had to happen eventually. You can’t get through this stuff without having something like that.
Mike [00:07:26]: Right, and these bugs are things that probably would have come up anyway. I would expect that the two we specifically ran into are ones that we would’ve come up with across other customers as well – eventually. Maybe not tomorrow, or the week after, but it would’ve been soon, and I’d rather find out now than later.
Rob [00:07:41]: So, what are we talking about today?
Mike [00:07:43]: Today, what we’re going to do is go through a bunch of listener questions that have come in. If you’re interested in having us answer any of your questions, you can email them to us at questions@startupsfortherestofus.com. Our first question comes in from Dan Gravel, and he says, “Hi, guys. I’ve been listening to the podcast since its launch, so thanks so much for all the tactical, actionable and practical advice you give. I’ve been running my business for almost seven years, and I’ve been full-time on it for six. The first product, called ‘Bliss,’ which you can find at www.blisshq.com, is still a bestseller, but it plateaued a few years ago. It’s a B-to-C, downloadable software product which automates the management of large music libraries. Arguably, the reason for stagnation is a drop in interest in self-stored music collections with the mainstream move to streaming. I’m considering moving all or most of the app into the Cloud and adopting a subscription payment model. I can’t move it 100 percent because a small software agent will always be required to perform the work on the music files. I think moving to the Cloud should lower friction in on-boarding, allow easier life cycle emailing and a higher LTV due to subscription payments that may enable a paid acquisition. How would you go about deciding whether to go ahead with this? Dan.”
What are your thoughts on this, Rob?
Rob [00:08:45]: Well, I think there’s two markets for moving to the Cloud. One is your existing customer base, and they’re easy because you can just ask them – right? You could do a survey, some one-on-one phone calls, and you could say, “Would this be of interest to you if we had a subscription version of this?” You can talk about the price point. You can talk about the benefits and find out if any of those folks are interested. Then the second market is everyone else – right? It’s all the audiophiles. You said you have an existing market. It’s some audio files, and there’s integration’s, and there’s certain customers who aren’t your customer yet, but who could potentially be. Those guys are a little harder to reach, but you could certainly survey your email marketing list. I’m assuming you have some type of list that is folks who’ve signed up to hear about updates, or maybe they follow some content you produce or something; and that would tend to be a much larger swath than folks who are actually paying you. That’s a good place to start.
[00:09:38] The other option is something I heard about from Patrick and Price Intelligently. He mentioned this in his MicroConf talk this year, and it’s a website called aytm.com. It’s AskYourTargetMarket.com, and you basically define all these demographics. I would imagine by this point you have a pretty good idea of where these folks live and if there’s any type of gender bias, if more of them are men versus women; what they do for a living; how they think about stuff. You can basically just define this at aytm.com, and then you pay per survey responded, and you could basically present the product as it is today and ask if they’d be interested in a Cloud version and try to do some market research that way. So, that’s the most data-driven way that I can think of and probably where I’d start to at least start getting some insight into whether or not this is a good move for you.
Mike [00:10:24]: Yeah, I’d have to agree with you that going back to your existing customers and asking them whether or not that’s something that they’d be interested in is probably a better bet, especially if you’ve been full-time and this has been the major source of income for the business for the past six years. The downside, I think, to doing that is that you have probably attracted a certain number of customers, or a certain type of customer because of the fact that it’s one-time, downloadable purchase and they can just buy it once, install it. Then they don’t have to worry about it ever again. There’s a certain profile of person you’ve probably attracted because of that, so I think that the data is probably going to give you at least some mixed messages there, because a lot of the people who fit that particular profile are not going to want to pay for a subscription service – not unless you can come up with solid justifications for what your service is going to allow them to do. Obviously, those are things that you’re going to have to work with those people to figure out what is it that they actually want and what would the Cloud service really do for them.
[00:11:18] One thing that I can think of off the top of my head is to be able to stream their music to any of their devices from anywhere; but then, of course, you’re going to rely on being able to take their own music and then replay it back to them through the Cloud, or through a streaming service of some kind that you would have to offer on the backend. I would imagine that all that’s possible. It’s just a question of whether or not the people actually want that and whether or not the existing services through Pandora, or Amazon, or any of the other ones already overlap enough with what you’re doing to be able to replace it and serve as a solid competitor; because they’re going to be heavily funded and already have access to a large market of people, but those types of people are probably not the same ones that are in your market.
Rob [00:12:00]: The nice part is that there are so many B-to-C services that are paving the way for you. As we know, B-to-C tends to be a tougher market, because you’re going to have lower price points, higher support, just a lot of things that aren’t as ideal with B-to-B market; but Apple, with iCloud and with iTunes Match and the Spotify subscription stuff; people are used to Netflix and Hulu. There’re just so many more subscription services than there were even two or three years ago, and consumers are getting more used to paying for these. I do think there’s at least some precedent for you to ride on the coattails as folks are getting used to more of these subscription pricing structures.
Mike [00:12:40]: So, Dan, hopefully that helps answer your question.
Our next question comes from Zachary Kesson. He says, “Hi, guys. I’m starting up a SaaS product and having a problem. I don’t have any traffic. My best day, I had 31 visitors to my blog, but they stayed an average of seven seconds each. It seems that all the sales advice I hear for SaaS founders assumes that the founder has a mailing list, and I don’t have one; and without some traffic to my webpage, I don’t see myself creating on in a realistic timeframe. At this rate, I should have a 5,000-person mailing list by September 2031, or something like that. I’ve put links in Reddit, tweeted them, put them in LinkedIn, but nothing seems to generate more than two to four clicks. Paid advertising is outside of my budget right now. What would you suggest? Zach.”
Rob [00:13:16]: This is saying, “How do I market a SaaS app with no audience?” The answer is you don’t want to. You want to have had a landing page up since you had the initial idea and to have at least been talking about it for a longer period of time so that you get some interest. Even an email list of 50 people is incredibly powerful at this stage. You don’t need 5,000, because having 50 people will allow you to get feedback and to do some type of customer research and figure out who wants to use it and for what, and talk to them about value propositions. That is much more valuable. I think that the fact, Zach, that you haven’t done that yet really puts you in a tougher position; because now you have to start from a dead-cold stop, and you’re saying paid advertising is outside your budget right now.
[00:14:11] AdWords is obviously very expensive, so you’re not going to do that, but you can get super-cheap clicks on something like You Tube; or, in Facebook you can get 10- and 20-cent clicks if you’re doing targeting. So, I would instantly put the Facebook retargeting pixel on your site even with 500 visitors per month. You said your biggest day was 31 visitors in a day, so if you do the math and it’s 900; and I’m assuming, since that was your biggest day, you don’t get that many every day. Even with 500 uniques a month, you can start some retargeting. I would also put an email capture widget on your blog. You’re not trying to build a massive list to market to, but you’re just trying to get some human beings that you can talk one-on-one with. You could test that versus, like, a Koalaroo survey box, or one of those other types of things. You’re trying to get information why are people leaving after seven seconds, because you have a problem if they’re leaving after seven seconds. Either your current traffic sources are garbage, or you’re not a fit for what people expect when they come to the site.
[00:15:08] You’re not going to get [out of this?] with split testing. This has to be a one-on-one effort, because you’re so early in the game. The fact that you don’t have a mailing list – and, again, even a mailing list of 50 or 100 – it puts you behind the eight ball. You have a product now, and it’s not going to market itself. You’re going to have to hustle. These days, you’re going to have to do everything under the sun that doesn’t scale just to get people using this app. So, putting a link on Reddit or tweeting it when you have 100 or 200 followers – that isn’t going to cut it. There’s too many people doing this these days to really be able to just do it with this kind of cold, random traffic. So, I would try to do a lot of one-one-one stuff to figure out who’s coming, why, what they expect, why they’re leaving quickly. That average number of seven seconds per visit is worthless. You don’t care about that, because a bunch of people are leaving after one second, and then some people are leaving after hundreds of seconds, and you want to focus on the people who are staying for those minutes. So, figure out how to differentiate those, whether you use Mixpanel, or whether you use Kissmetrics, or whatever it is you use – free trial of Crazy Egg, or Inspectlet. These are things that can show you what people are doing on your site, but don’t rely on the tools to do it. You’re going to have to dig in and do a lot of one-one-one stuff.
[00:16:17] The last thing I would say is if you have an idea of who should be buying your software, figure out where are they; because being on Reddit and Twitter is too generic. Using your tool relates to other CRMs, it looks like, so I’m assuming you integrate with something like maybe Salesforce, or Closeout iO, or [Bay?] CRM, or whatever. So, how do you get into those integration repositories? How do you get those guys to co-promote? You promote on your end when you finish integrating, and they co-promote to their audience via a blogpost. That’s integration marketing. Ruben Gomez was the first person I ever saw do it, and then I coined this term “integration marketing” and talked about it at a MicroConf talk a few years ago. It’s to get marketing done via these integrations, and in order to do that you have to hustle. It’s not going do it on its own. You need to communicate with them and convince them. You have to have some type of blog at that point if they’re going to blog about it, because you have to have something to offer them. They’re not going to email their list if you don’t email yours. Since you don’t have that, you don’t have that asset. An asset doesn’t build itself overnight. This is stuff that you need to be thinking about well in advance of when you need it. If you haven’t to date, then today is the day to start building that. It’s going to take you longer than if you’d gone about it in the correct order, or in a more optimal order; but you can’t really look back at this point. I think you’ve just got to get started building these assets as of today.
Mike [00:17:39]: I went over and took a look at his site. It’s yourcrm.link. It looks to me like the product itself is aimed at SaaS businesses who want to connect with CRMs, and that was just my initial idea of what the product does based on what I read there. If that’s the case, what I would probably do is move much more towards an [out?]-[?] model where you’re directly contacting those people that you think are going to be a good fit. So, sit down and make a list of the 25 companies that you think would be a good fit for using your product and then contact them. You just go through the list and talk to every, single one of them just to get a yes or no as to whether or not they’ll talk to you and see if it actually is a good fit. It sounds to me like if you’re not getting targeted traffic to your website, then you’re not posting things in the right places. If you don’t know what that target is going to look like, then it’s going to be very difficult to figure out where you should even post it. Is Twitter a good place to be posting those? I don’t know. It really depends on what the people who are interested in your product have to say and where they actually hang out. I can’t answer that, but I think that you should be able to after you have some of those conversations, and I think that’s the bottom line at this point. Because you haven’t had those conversations, it’s very difficult to figure out where you should be marketing your product to and who is an ideal customer for it.
Rob [00:18:55]: Yeah, that’s actually a really good answer. I think I may have liked your answer better than mine, because yours really is doing the one-on-one work and the outreach. It’s just a matter of getting in these one-on-one conversations and figuring out who can use it and why would they. His price point starts – I hadn’t even noticed that part, because I didn’t scroll down to the bottom. His price point starts at $149 a month and goes up from there. It’s like $149, $499 and up, so there’s definitely enough room to do some medium-touch sales and some high-touch stuff. I think that’s a really good start. The other stuff I talked about is probably as you want to drive more traffic down the line. I think these one-one-conversations are more critical up front.
Mike [00:19:34]: Yeah, and I think your point about not worrying about trying to do any sort of split testing or anything like that is also valid because of the fact that you just aren’t going to have enough traffic to be able to do that. It’s those one-on-one conversations that are going to be the single most valuable thing to you at this point. What I would probably recommend is visualizing what your sales funnel is going to look like just from a conceptual standpoint. You don’t have to be exact, but you could just set it up as a simple three-, or four-, or five-stage sales funnel where you say, “These are the 25 people that I would like to see as customers,” and they’re at stage 1. You haven’t even actually reached out to them. Then stage 2, you’ve reached out them. Stage 3, you’ve had a conversation or scheduled a meeting. Step 4, you’ve had the meeting; and then Step 5 is whether they are actually going to trial the service or sign up for it. At that point, if they pay for it, or they’ve decided that they’re going to evaluate paying for it, then they drop off that particular sales funnel right there; and then they maybe enter into another one.
[00:20:31] Those are the basic steps that you should probably go through. If you’re just doing things quick and dirty because you don’t have any of this traffic or any stats, just set up a Google doc or something like that, or a series of them. Every, single conversation that you have, you put that conversation into the Google doc. If you have a second conversation with somebody, maybe because they’re further down on your sales pipeline, you add that conversation into that Google doc. That way, you end up with a series of four or five different Google docs that are in line with what your sales funnel looks like, and you can always go back and you can refer to all the conversations that you had at stage 1 of your sales funnel, then at stage 2 and stage 3. It lets you look back at those down the road to see if there are any conversations that had a lot of overlap, or a lot of questions that specifically came up at a certain stage of your sales funnel. Then you can take that stuff and translate it back onto your website to help gather trust and answer questions for people that you are not directly talking to. So, that’s essentially the customer development process that I would go through for this.
Rob [00:21:30]: And if you need software to help you do that cold outreach, I would look at Bluetick.iO.
Mike [00:21:36]: [Laugh] Ah, yes, thank you for the plug there. So, Zach, hopefully that helps you out.
Our next question comes in from Corey Moss, and Corey’s been a long-time listener of the show. Corey left a message on our voicemail number, so we’ll play that for you now.
Corey [00:21:47]: Hey, guys. This is Corey Moss. Over the last year, I’ve stepped back from SaaS apps to concentrate on my Kanban board plugin for WordPress. You can check it out and the paid add-ons at kanban nwp.com. In the SaaS world, the holy grail is monthly recurring revenue, but in the WordPress ecosystem, most licenses are annual, and most don’t even auto-renew, if you can believe it. I want to apply SaaS best practices, but I’m nervous about doing things too differently from how people expect it. Mike, have you run into this in dealing with enterprise? Have either of you had experience with this disrupting the status quo when it comes to monetizing patterns? Thanks.
Mike [00:22:25]: Corey, if I understand you correctly, you’re essentially asking how to disrupt the current payment that a market is used to in standard practice today. So, your question to me specifically was did I run into that sort of thing in the enterprise market. Yes, I did. It was very difficult to get them to change their ways and, looking back on it, it was probably foolish of me to even try, and I would not recommend that somebody else try to do that. So, if you’re going after the WordPress space and you’re trying to convince them to pay for a subscription to something that they’re used to paying for once as a downloadable product, and potentially a yearly maintenance fee or something like that after that, I would be very hesitant to say that you should try and modify that model in such a way that they are not comfortable with or they’re not used to.
[00:23:10] I’ve talked to other people who are in the WordPress space, and they said that selling subscriptions in WordPress is a very difficult way to go, because people are just not used to it, so they don’t do it. I think Rob had one of his yearly predictions that this is going to change in the future, but I think that there’s also going to be a huge subsection of the market that is not willing to change, and it’s going to be years before they are convinced that that’s the only way to go because they are forced to do that. Even if you look at today, it’s 2016, and there is still a ton of downloadable software that’s sold out there. What we tend to see is mostly the recurring revenue, but there are huge number of downloadable products that are still available for sale; and people still buy them every, single day because they only have to pay for them once. I think that you’re probably going to run up against that roadblock in the WordPress space for several years to come. If I had to put a number on it, I’d probably say between five and ten. So, I would not bank on trying to do that overnight, and I almost certainly wouldn’t even recommend going in that direction if you can help it. I would probably double down on trying to figure out what people are willing to pay more for and try to increase your lifetime value that way as opposed to trying to change the way that they fundamentally buy their software.
Rob [00:24:19]: I know that there are some knowledgeable WordPress folks who are trying to attack this problem, and they haven’t really been able to break through yet. I don’t know of any WordPress plugins that don’t rely on an external service that are able to pull off a monthly pricing tier, because if you do have an external service – let’s say it’s backup to a Cloud server, and you’re maintaining the Cloud server, or it’s some other thing like – I think Jetpack doesn’t have external stuff that WordPress runs. They’d be able to charge a monthly fee. But in terms of just doing it for the sake of doing it and not having a strong justification? I agree with Mike. I think it’s going to be an uphill battle.
[00:24:53] I think the other thing to think about is, while monthly subscription is the holy grail of SaaS, that’s also its biggest drawback. It takes you years to get through the “long, slow SaaS ramp of death” – right? It’s just years of toiling away to build up any type of monthly, recurring revenue. The nice part about WordPress plugins is you can charge more up front. You do get a one-time fee. Let’s say you’re charging 49 bucks for the annual license, which a lot of people are doing. I think auto-renewing with advance notice – let’s say a five-day, or a three-day, or even a week notice before you auto-renew should be perfectly acceptable. I do know that some folks are moving in that direction. I think the more savvy WordPress business folks are doing that, and I would as well. The nice part is that, since most of the traffic that you generate will come from these recurring sources like a Google, or a WordPress.org plugin repo, you can ramp up your revenue to a couple grand a month really quickly; whereas SaaS is just going to take forever to get there. While SaaS will grow over time, I’m not so sure that’s what you want at this point. I personally would be going for that early revenue. Let’s try to get to between 2 and 5 grand a month as quickly as possible, and then you can start thinking about either adding add-on services that could potentially become subscription; or, using that money to stair-step your way up with either a second plugin, buying out all your time and then thinking about something long-term that is more of a monthly subscription model. Whether that’s in the WordPress space, or you just go straight for SaaS, that’s something to think about.
[00:26:21] But I am not bullish on the fact that, as a relative newcomer to the space, that you’re going to be able to come in and just charge monthly without a major justification; because so many WordPress users, the reason they’re using it is because it’s cheap and because there aren’t subscription fees. A lot of consultants use it because they know that their clients don’t want to pay subscription fees for things. They don’t want to pay for Shopify or Squarespace. So, if you come in and want to charge this 9 bucks a month or whatever it is that you could get for it, it’s going to be a deal breaker for a lot of folks.
So, that’s my advice as of now, the middle of 2016 here. I do believe that over time this will change. I think most things are going to move toward subscription. We see it happening in B-to-C entertainment, which was unfathomable a few years ago to think that millions of people would be paying for Spotify, for essentially music subscriptions; and even 15 years ago, to think that tens of millions of people would be paying for Netflix, for movies by subscription; but it’s happening. Do you think that WordPress will eventually go? I think it’s years out, and I think you’re ahead of the curve.
[00:27:22] So, my prediction Mike called out. As we’re talking through it, I’m realizing there’s just no chance that’s going to happen in this year. I think what my prediction was was that one person would figure it out pretty well; and that, of course, remains to be seen. I guess we’ve got another six months.
Mike [00:27:35]: I’ve seen one company that has done this, which I think they’ve done pretty effectively, Optin Monster. They have a little widget that you can install on your website, and they do charge a subscription fee for it, but they also have – as you pointed out, unless you have some sort of external service that integrates with, like for WordPress backups, or security checking, or things like that, where it needs that external service in order to function, and that external service needs to be up and running at all times, maintained and everything else. Optin Monster can do that because, one, the types of things that they integrate with are types of services that you’re going to pay for. So, it really sits on the front of that funnel and is aimed at those people who are used to paying for those things.
[00:28:16] The other thing is that I think they charge $50 or $100 per year, and it is that subscription fee; but they also keep the widgets up-to-date. So, if there are changes made in MailChimp, or Drip, or Aweber, or whoever, they will update the plugin so that when it comes down to your site, it is all up-to-date and you don’t have to worry about keeping your own plugins up-to-date. So, there are some advantages there, but they also made that switchover after they hit, like, 200, 300,000 customers or something like that. They’re not going from ground zero and trying to build a SaaS pricing model out of it. They already had a pretty substantial user base before they made that change.
So, Corey, hopefully that helps answer your question.
[00:28:54] Our last question for the day comes from Brian Kenry, and he says, “Hi, guys. Love the show, and it’s been a big help in getting my startup off the ground. I’m building a platform that will be monetized by generating leads and selling them to a private agency. I’ve assumed that I will partner with one agency and turn over leads at a cost per lead, but I’m wondering if I should consider a different dynamic. I realize this is a vague question, but is there any benefit to pursuing a subscription model or a marketplace for leads? It seems like a cost per lead is the most logical way to approach this, but I want to make sure I’ve considered all avenues. As always, thanks for your input.”
Rob [00:29:21]: I think I would stick with cost per lead, because that’s the most granular, and it’s the easiest to understand both from your and your customer’s perspective. Your costs are going to be variable, but you’ll be able to lock them down below a certain amount and then know what your margin is.
One thing I would consider is having customers – they could pay on a subscription basis so that they pay in advance. Here’s what I’m thinking. Instead of delivering a bunch of leads to them and having them pay at the end of the month or something, you really want to get that payment at the start of the month, knowing that you’re going to deliver those leads over time during the month. So, you could think about having tiers of, like, “Do you want 100 leads? Then it’s 1,000 bucks,” or whatever. If you want 200 leads, then it’s either $2,000, or you can give them some type of discount. Do $1,800 or something a month. The nice part about that is you don’t have to chase after them for money. They can either sign up for the PayPal subscription, or through Stripe, or whatever; and you’re automatically billing them, and you’re getting paid in advance. If their payment doesn’t go through, then you don’t deliver the leads. You’re not out the money. I think that is beneficial for you from a business perspective, and it’s going to save you time. That still works out to a cost per lead, but they’re basically committing up front to buying a certain amount of leads, and you’re getting that payment up front.
[00:30:34] Those are my thoughts off the top of my head. I can’t think of any way to get more complex without just making things hard to understand, and I know in the lead industry things are typically sold per lead. Do you have any other thoughts on it, Mike?
Mike [00:30:46]: I have a few that go into different places on this. As you said, the one that’s easiest to understand is just selling them directly cost per lead. The problem there is that you don’t know how much time and effort it’s going to take you to do the work to get a lead. As you said, if you do it enough, you’re going to able to figure that out and you’re going to be able to do some averages there and then tell people, “Yeah, I can get you X number of leads for Y dollars.” Then eventually, they’ll be able to make that mathematical calculation in their head or just look at it and say, “Yeah, that totally makes sense for us to do that.”
The other thing that I can think of that might be an option is to have them pay you for a specific service based on a flat fee. Whether it gives them results or not is a risk that you’re pushing on them. I don’t like this idea nearly as much because of that very thing, but you could say, “For $500 a month, we’re going to do X, Y and Z for you. We’ll reach out to X number of people with your emails, or with your script,” from whatever the call is, whether you’re doing cold calling or cold emailing. You can do something like that and put together different packages.
[00:31:51] I do agree that you should probably charge people up front so that you have the capital on hand to go do that, so that you’re not paying for or funding it up front only to have somebody cancel later on, and then essentially eat the costs associated with that. I’d definitely collect the money up front. Those are the things that come to mind for me. The real big there, I think, is who is assuming the risk for doing this. I think that you probably want to assume as little of the risk as possible, but I also think that you probably don’t want to go down the road of saying, “We’re going to do X amount of work for you,” but then basically push all of the risk onto them if it doesn’t deliver; because it could reflect very poorly on you if you go through a month, or two months, or something like that and, for whatever reason, you’re just not getting the results that you used to. I think that you’re better off evening those out for the customer and letting them look at the raw numbers and say, “Yeah, we expect to pay $10 a lead, and that’s what we’ll get.”
Well, Brian, I hope that answers your question.
Rob [00:32:45]: And that wraps up our episode for the day. If you have a question for us, call our voicemail number at 888.801.9690, or email us at questions@startupsfortherestofus.com. Our theme music is an excerpt from “We’re Outta Control” by MoOt, used under Creative Commons. Subscribe to us on iTunes by searching for “startups” and visit startupsfortherestofus.com for a full transcript of each episode.
Thanks for listening, and we’ll see you next time.