Episode 288 | How to Choose and Test a Paid Marketing Channel

Show Notes

In this episode of Startups For The Rest Of Us, Rob and Mike talk about how to choose and test a paid marketing channel. They give you some criteria and discuss the mindset you need for choosing the channel that’s best for your business.

Items mentioned in this episode:


Rob [0:00]: In this episode of “Startups For the Rest of Us” Mike and I discuss how to choose and test a paid marketing channel. This is “Startups For the Rest of Us” episode 288… 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 have built your first product, or you’re just thinking about it, I’m Rob.

Mike [0:29]: And I’m Mike.

Rob [0:30]: 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 [0:33]: You changed up the intro a little bit.

Rob [0:34]: I know, I added in some more stuff. Because we always talk about launching in the intro, but really, now often times we’ll talk about the building, and the growing, and the scaling and all that. Maybe I’ll update that actually in the doc. That was, kind of, off the top of my head.

Mike [0:48]: Well, of course, you know, there was the little flub there, which will thankfully be edited out.

Rob [0:53]: Indeed. So what’s going on?

Mike [0:54]: Last week, I had talked a little bit about how I had started onboarding people into Bluetick, and I recently just uploaded a spreadsheet into Bluetick so that I can have it automatically send out the emails to people to get them scheduled, so I don’t have to individually email each person. And of course the nice thing is that it will gradually email them out over time instead of all at once. So that’ll be helpful in terms of putting a little bit of spacing in between some of those. Obviously, they can choose anytime on my calendar, but it will be nice to not just use it in kind of production environment, but to also let people who are coming on to it see the app in action and what it will look like on their side of things.

Rob [1:31]: Yeah. That’s pretty cool. So, you’re at the point where you’re able to dogfood it, and use it for your own use.

Mike [1:36]: Yes.

Rob [1:37]: Yeah. You know, I remember when we first hit that moment where we could do it with Drip, and we started really using it for a lot of our email marketing. At that point it was like something flips, and you’ll suddenly — you already look at your product a certain way, because you have been building it. But something, kind of, shifts once you start using it, because you’re going to notice everything that’s clunky about it, everything that you need to fix. Even more so than you did before you started using it. So I think it’s really a plus to be able to use your own product, and to catch bugs before the people – or even to catch usability issues before the rest of your customers do.

Mike [2:11]: Yeah. I mean, I started trying to do this a couple of weeks ago, and as you said, it was clunky, and there were things where I ran into where I could do those functions and perform those actions, but it was a little painful. There was a few too many clicks, and there’s still a few places where we could optimize what we’re doing in the user workflow, and how they get to different places in the app. But we basically redesigned the entire navigation, and merged a couple of pages so that people don’t have to click around as many places. It’s a lot better today than it was even just a week or two ago. We just pushed out a new version of it this morning. It’s nice to be able to do that now, and as I said, it was working a few weeks ago, but it was a little painful. I think we are starting to turn that corner where the app – it’s not just inherently useful but – it’s nice to be able to dogfood it in a way that is not as painful.

Rob [2:59]: Yeah, for sure. I think there’s that quote that’s kind of over-quoted at this point about how, “If you’re not embarrassed by the first version of your app then you waited too long to launch.” I think that there’s two sides to that embarrassment. One, it can either be that you just don’t quite have the usability right yet. I think that’s one reason to be embarrassed. I think another one is, maybe it is really useable, but you just don’t have enough features to do what you want to do, and you feel like it’s sorely lacking. I think those two axes can either be off on one or both of them. But identifying which one, and kind of moving up market, or even just improving upon those axes, I think, is a big thing in the early days.

Mike [3:33]: What about you, what’s going on this week?

Rob [3:34]: Well, we had a pretty good response to a recent blog post on the Drip blog. It’s probably the post that’s gotten the most uptake in that last several months, and it was called, “Email Marketing Tips: 33 Industry Experts Share their best Strategies and Tactics.” It was an expert roundup, basically where really Zach on my team headed it up and did most of the legwork, although I contacted some of the people who I had contact with. But we got some really solid tactics from people like Devesh Khanal from Grow and Convert. We got Josh Pigford at Baremetrics, Joanna Wiebe at Copy Hackers, Dan [Norestelli?], Alex Turnbull, Brendan Dunne, Patrick Mackenzie, Noah Kagan – just this big expert roundup, and there’s some really interesting tips in here that people are using to improve open rates, and click through rates, and just overall just improve stuff. So it went to the top of GrowthHackers for a while. I think it went up to one or two on inbound.org, and at this point it’s got 300 social shares. If you haven’t checked it out, it’s a really good post, and I’m not just saying that because we put it out. There’s a lot of cool tips and tricks in here. We’ll obviously include that link in the show notes, but if you go to the Drip blog, or if you just search “33 Industry Experts Best Strategies and Tactics” you’ll find that.

Mike [4:44]: Awesome. So what are we talking about this week?

Rob [4:46]: Well, we periodically get questions from listeners about getting started with paid acquisition, and today we are going to be talking about how to choose and test a paid marketing channel. Most recently we received this question from Ryan Frank, and he said, “I’m sure this had been requested multiple times but I’d love to get some insights on the show regarding setting up and running marketing campaigns. What platforms do you choose to advertise on, and why? How much is a reasonably initial investment? How do you brand the campaign? How do you target? How many do you run at a time, etcetera?” He goes on to ask some more questions that we’re going to cover today. The interesting thing is we don’t tend to cover this topic specifically because—well, there’s a few reasons. One, this stuff changes so frequently. Even the ad networks that are going to work for bootstrappers will completely roll over in, let’s say, 12 to 24 months, and anything we say here about specific ad networks will be different then. In fact, with Facebook – for a while there – it was like every 90 days things were changing in the ad interface itself, and different ad types started performing better than other ones. You just had to be right at the cusp of it, and either in touch with someone who really knows what they’re doing and are constantly testing, or you have to listen to some really detailed, recent information. That’s why we don’t often dive into specifics. However, today we are going to talk more about like the mindset, and the approach, of testing and choosing a paid marketing channel. Another reason we haven’t talked about this a lot is it can get often to the weeds, and it can get a little too detailed and boring. We’re going to try to avoid that today. Then, finally there’s really this, kind of, factor of competition. And this has become more of an issue of late, with competitors kind of, I’ll say, borrowing heavily from some approaches, which has been a bummer, and it comes with a little but higher stakes now that I have ten people working for me, and we’re paying mortgages and doing that kind of stuff. So, kind of, giving away exact detailed information of exactly the wins that we’re having, and why, isn’t necessarily as feasible as it used to be when I was selling eBooks and beach towels. With all that said though, there is a really specific path that I take when I’m thinking about testing and choosing paid marketing channels, and picking out new networks, and how to test. And we really are going to dig in. There’s a lot of meat in today’s episode.

Mike [6:53]: Awesome. So, let’s dive right in.

Rob [6:54]: Yeah, so we’re going to, kind of, break it up into two sections. The first is the criteria for choosing, and the second one is really how to run a test on a paid marketing channel. The interesting thing is, it’s only getting more complex these days choosing a platform to test on. There are more choices, and within those choices there are more choices. There weren’t YouTube ads at all, and then there were YouTube ads. And now there’s YouTube Instream, and there’s YouTub- in-display, and there’s another type that I forget the name of it. And with AdWords, it used to just be the little text ads, and then there’s the display network, and then there’s the text display network, and the visual display network, and Gmail ads, and on and on and on. It’s just getting more and more ubiquitous. The nice part is that does give you more choices and more opportunities, but the negative is it makes it hard to choose. I think the kind of first thing I’ll throw out is that, these ad platforms, there the cheapest when they first launch. That’s also when they’re hardest to use, because the management tools are not very good, and there’s not a lot of insider info on how to, kind, of use them for optimum results. So you really are pioneering, and you’re going to just do trial and error. But as these ad platforms get more advanced, and they get more matur – like AdWords is an example – they become so expensive that there’s just no chance that you’re going to make them work. Whereas a platform like Facebook is probably in the middle right now, it used to be cheap. It’s not totally out of the range of being affordable, but it’s in the middle. Then somethings that’s really new – maybe Pinterest for example, or Instagram’s pretty new in terms of ad networks – those will be the cheapest clicks, but there are also maybe the hardest to use, hardest to target, and you don’t have a ton of information on how to do that.

Mike [8:29]: I talked to somebody – probably a year and a half, two years ago – about some strategies that they were using on Instagram for paid acquisition, and they showed me a graph and I looked at it. It was insane. They basically doubled their revenue almost overnight by doing Instagram marketing and just paid acquisition through that channel, and there was no mechanism for doing it. They basically built an entire system themselves. They had all these spreadsheets and data, and they were calculating all these different metrics on their own. And for themselves, they pioneered it. They didn’t really talk about it publically, but I looked at what they were doing and it was absolutely incredibly and insane. The results that they were getting, it was really, really hard to argue with the results, just because there was a point and time where they weren’t doing any of this type of advertising, and then they started, and immediately you could see the graph. It just doubled the revenue almost overnight. You are right, I think, about getting in when those platforms are new. I think the downside of that is that when these new platforms come out you can spend a lot of time and effort trying to get something working and running, and it may just not work. Because that’s one of the risks that you’re taking with those, because you are pioneering something completely new, there’s no benchmarks to go off of, there’s no guidelines, there’s really not very many people you can talk to, or ask about, in terms of being able to work on those channels. But if you can find something and make it work, it can be really, really cost effective.

Rob [9:51]: Yeah, that’s pretty cool. I remember that guy’s story, assuming it’s the same person that I’m thinking of. He basically went in before there were even official ads. There was no ad platform, and he was paying the more popular Instagram users X dollars to post the thing. It was in essence, ads, but he was so early that he got in when it was really cheap. I heard Noah Kagan talk about this – before advertising in email marketing newsletters was getting really big – he would just approach big email marketing newsletters and make them an offer. That’s a kind of way to get around the ad networks, because the ad networks are marketplaces, right? And eventually they kind of even out, especially once your competition is there, then your cost per click is just going to go up and up because there is competition for those clicks, and your cost required is going to be similar across competitors. So it is going to even out eventually at some type of norm, but if you get in way early then that competition doesn’t exist yet, and you can get the cheapest clicks early. There’s obviously a lot more danger there. That’s really the balance.

Mike [10:46]: Right. I guess one of the takeaways from that, that I just thought of off the top of my head, was that if you are getting in that early, and there isn’t a marketplace for ads on a particular channel then you could probably negotiate for some sort of exclusivity when you’re working with those people. And if you can get it, then you can, sort of, lock your competitors out of that as a channel. That’s something that also worked extremely well in that particular case. That’s something you might be able to take away from that.

Rob [11:14]: I think another question you want to ask yourself is, are you going to do this yourself? Like, are you going to test the channel yourself, or are you going to hire someone? This depends a lot on whether you want to learn it or not, whether you have time to learn it or not, and whether you have budget to hire someone. The nice part is, three or four years ago, hiring someone to do it was very, very expensive. These days there are knowledgeable folks who are managing AdWords, or Facebook ads, or Twitter ads, or I’m sure Pinterest ads – I haven’t used anybody for that – but that are not these big agencies that charge enormous sums of money – or a percentage of ads spanned – but that may charge per lead, or that just charge a lower monthly fee. There was someone that was going to manage Facebook ads and I didn’t wind up using them, but it was a few hundred dollars a month rather than several thousand like a lot of the old agencies. I think, keep in mind that running a test is not terribly hard, but it is time consuming to actually get this to work and then scale it up. So be realistic about how much time you actually are going to have to run these ads. To take a step back, I’m realizing we didn’t really cover when you should try to do this. Like, when should you think about running paid ads. I think that the two points are: really early on when you are maybe testing a value proposition on landing page, and you want to send someone, and you want to test that value proposition and get email signups. I think it’s very valuable then. I think in the interim – kind of between that launch and product market fit – when you are just trying to get more people to use it so you can do customer development, I think that’s a good time to run it. You don’t want to scale that one up. You just want to get enough people in there to get enough feedback to improve your product. That’s the second time. The third point, I think, is once you’ve built a product and you have a value proposition and things are really starting to scale up, that’s when you want to hit this really hard, and I would recommend actually getting the budget together and hiring someone. I think in the first two instances you only need such a trickle of traffic when you’re testing stuff, that I would guess unless you do have money at your disposal – like you’ve raised funding – that it’s not going to be worth going out and hiring someone for the small amount of clicks that you’re actually going to be buying.

Mike [13:10]: This is one of those situations where, as a bootstrapper, you’re kind of caught between a rock and a hard place, because in order to spend the time and effort to get the results that you want, it’s going to be time consuming, and you either have time or you have money; you typically you don’t have both when you’re doing this. It makes it painful to go in and start running some of these experiments. You have to pick and choose your battles a little bit in terms of where you’re going to try and optimize things inside of your ad campaigns, because it’s going to either cost time or money, and you typically don’t have both available to you. So often times you are trying to implement things, or run some tests, and they’re going to be less than optimal. I think that that’s an important thing to keep in mind, is that you’re not necessarily looking to optimize each and every single test. You’re looking to learn from the early ones so that you can scale things up and make them better in the long run. You don’t want to spend a lot of time and effort trying to figure out, “Okay. What is the single best ad that I can come up with?” Take three to five of them, throw them against the wall, see what happens and then measure the results and iterate from there. It’s not about getting things right on the first time. It’s about iterating over time and then enhancing your results.

Rob [14:20]: Another criteria to think about when you’re choosing an ad platform is, is it B to B, or B to C, or is it a mix? I’ve found that there’s a lot more B to C inventory than there is B to B. The B to C inventory tends to be really low quality, and so it can work, the clicks are really cheap, but if you’re not targeting something like weight loss, or online dating, or these really broad consumer interests, it can be hard to make money with those more B to C networks. And those are things like Chitika and Clicksor, and I think Infolinks is like that. I tested 10 to 15 ad networks back in the day when I was really trying to scale HitTail, and I found that most of the ones that I hadn’t heard of were just kind of junk traffic. They were super cheap clicks – five cent, ten cent clicks – but average time on site was two seconds or three seconds or something. So it was pretty noticeable that either the clicks genuinely were just bots or not real people, or they were accidental things, or they really were just looking for more stumble-upon content to hang around and be enticed with info market stuff, rather than actually trying to sell a product. You definitely want to think more. We’ill talk through a few networks here. My next bullet spells out several networks, and there still are a lot of B to B and mix networks that do have quality traffic, but you’ll want to think about this in terms of the product you are selling, and the type of scale that you’re trying to get to.

[15:43] The next criteria to think about is, does the ad network have enough volume to scale traffic? Early on, like I said, when I was running ads for HitTail, I actually ran ads on Bing and Yahoo and other search engines, and those were examples of things that converted for me, but I couldn’t get enough clicks that it was even worth managing the ad platforms. So, while you can look at all these kind of peripheral ad platforms, it often will be a waste of time even if you can get it to work. The majors are things that you’ve heard about. There’s Facebook, there’s Twitter, there’s Google AdWords. Then there’s the AdWords display network, which is basically what AdSense runs on. Then there’s YouTube with multiple different types. There’s a LinkedIn ads. Pinterest now has an app and network. AdWords has a Gmail ad section that can advertise on Reddit, and I got those to work for a short time. BuySellAds.com is a nice display network, and then we start moving in to the consumer stuff. I think there may be one or two more that are at scale and you can do B to B, but the ones that I have already listed are pretty much what I know about. Then there’s like Outbrand, and Chitika, and Clicksor, and Infolinks. There’s a good article summarizing these. It’s on monitizepros.com and we’ll include that in the show notes. The idea here is, if you can find an outlier ad network and you can make it work, the clicks are going to be super cheap and it’s going to be very, very lucrative for you. The trick is, I’ve never been able to do that, and I spent quite a bit of time and money trying to do that at one point, and I found that the ad networks today that people are talking about, there’s a reason they are talking about them. It’s because that’s where the real people who actually spend money are. So when you think of the Facebook, Twitter, the Google, the LinkedIn, and the BuySellAds, and maybe Pinterest and Instagram if that’s kind of your thing, those are going to be the mains that you’re going to have to evaluate and think – based on what I am selling – is someone going to buy email marketing software on Pinterest or Instagram? The odds are probably not. That’s not going to be at the top of your list if you are going to advertise. You’re going to want to stick more to an ad network where you feel like you at least have a chance of someone being interested in your product.

Mike [17:47]: You know, the fact that you had an ad budget for Yahoo means you probably could have spent enough money to just buy Yahoo outright of what they were worth.

Rob [17:55]: I know. Interesting. For real. This is back in the day. They were still worth something a few years ago.

Mike [17:59]: Are they negative now?

Rob [18:00]: Oh, it’s so sad to see a giant like that just go down the tank.

Mike [18:08]: Oh, well. So, what’s next?

Rob [18:09]: So, the next thing to think about is to take into account what the cost per click is going to be, verses your budget, and the lifetime value of your users. You’re just going to have to ask around, or do Google searches or whatever, but these days we know that Google AdWords is very expensive because it’s such a mature platform and Google’s really smart. Whereas Pinterest and Instagram – since there are early – the ad clicks are going to be a lot cheaper. You can get some cheap clicks on BuySellAds, and I have heard there’s cheap clicks on YouTube these days. Whereas Facebook, like I said, is kind of in the middle. Twitter, I think, is in the middle as well. They are both starting to mature, and they are getting a little more expensive. So this is one where I think that the most value can be had probably in the early to the middle section there, and I think a lot of the older – or the larger – enterprises are doing the more advanced, or the more mature, platforms because they are easy and more qualified clicks, but they are definitely a lot more expensive.

Mike [19:00]: I think one of the things we might want to touch on here is talking about the cost per click, or the cost per lead, that you are getting verses your budget, and that kind of ties back into what the lifetime value for your customer is as well. So obviously, if you have a one-time sale for a product, then your lifetime value is going to be whatever the sale price of that product is. So, if it’s 50 dollars to buy a WordPress plugin, it’s going to be 50 dollars. But if you have a SAS product, or anything where there’s recurring revenue – and it’s let’s say 100 dollars a month, and people are going to stick around for 10 months – you’ve got a lifetime value of about a thousand dollars. I think that the general rule of thumb that I’ve heard from most people is that you don’t typically want to spend, on a long term basis, more than about a third of what you’re lifetime value is to acquire a given customer. Now, that’s not to say that you can’t spend upwards of break even for the first set of them, but you’re going to drive your business into the ground very, very quickly if you do that for too long. The primary reason for that is because you’re going to be spending a lot of your ad budget now, and let’s say that you spend a thousand dollars to get a customer now. But the problem is that you’re not going to make up that revenue for any given customer until they are there for the entire time period. So you’re not going to get all of that money back until 10 months into it, so that makes it very difficult to do that. When you’re trying to go through and budget for this, make sure that you are very aware of how much money it is that you are spending, and keep in mind that the tests that you are running early on, you may very well run way above and beyond that, and it’s primarily just to go through and do the testing on those different channels and iterate and try and make your ads better. You’re not going to get them right the first time. You’re going to spend a heck of a lot more money the first time through, but as you iterate, if you can get that under about one third of your lifetime value then you’ve got, essentially, a sustainable channel that you can start pumping money into. Now, that’s not to say that you can do it on an unlimited fashion, because even if you are doing that, then you’re still not getting that money for at least three months. So you can’t just pump every dollar that you have into it. You do have to be careful. You do have to budget for those types of things. I certainly wouldn’t run the bank account down to almost zero with the anticipation that you’re going to be able to make it up within 30 days.

Rob [21:12]: That’s a really good point. I think the one third LTV is a good reminder, and then – typically if you are bootstrapped – I see most people not going more than about three month payback. I see some people edging up to like a four month payback window on your trial or sale cost. So that’s not a click cost, but that’s someone coming in and actually becoming a customer, and you want to get paid back within four months. In your example, if you were doing a hundred bucks a month then you could pay up to 400 hundred dollars to acquire a customer. But, then like you said, you need the cash to be able to do that. Funded startups, the rule I’ve heard is, no more than a third of lifetime value, and no more than a one year payback. They tend to raise enough money that they can cover out, but when you’re bootstrapped — I remember first starting out it was like I had like two month payback with HitTail. Then as I got more cash it was three month, then it was four month. Then once I really optimized it, and I widened the reach, and I knew that not only were the folks signing up and becoming customers, but I was watching them remain customers, I actually kicked it up a little higher than that. That’s kind of the range that I would think about. I think the last criteria for choosing a network is, are you going to need banners or fancy images? And do you have a way to do that? I know Bannerarchitect.com is a pretty good way to do that, and there are certainly people on Upwork these days that can create banners. But keep in mind, there are some text networks that work really well, and you don’t have to spend the time and money to develop banners, and banners burn out as well, so you are not going to have them made once and have then last forever.

Mike [22:31]: That’s actually a really good point about the fact that those banner ads will burn out, and the same images that you use today will not necessarily work the same even a week from now. I mean, sometimes they burn out very, very quickly. Other times they will last for a while. Again, every image I’ve ever used – or ever seen – in any of my ad campaigns, they inevitably burn out after a while. And what will happen is you will notice that the effectiveness of that campaign will start to decrease even after you’ve optimized it. So you will have to refresh those images, and use different ones, and play around with it. And the thing that sucks about that is that it screws with your stats, because even week to week you can’t necessarily guarantee, or look ahead and say that, “Oh, this advertisement is going to have this effectiveness.” because it will change over time as people see it, and they basically become blind to it.

Rob [23:22]: So moving into how to actually test a network. As a rule, the higher your budget the faster you can test. But you can test cheaper channels with a budget of, say, $10 a day. You’re not going to be able to test AdWords at that budget, but you could start running let’s, say, a Facebook test, or even maybe there’s some cheap clicks – like I said – on YouTube, or Instgram, or Pintrest. I’m sure you could do those for 10 bucks a day. I personally like to start with about 20 bucks a day so we can move a little faster. And I commit maybe around a couple thousand bucks to a full test, unless it’s either doing really well at the start, or it’s really tanking at the start, we’ll go in one to two thousand bucks; that’s just the scale we’re at right now. But you could probably make a call after a few hundred dollars. I recall doing that. I mean, some of these are just going to be so obvious that they’re not working that you don’t need to invest a huge amount of money in order to figure it out.

Mike [24:10]: I think one of the things you have to be careful about when you are doing these tests is that you have to be aware that the way that the advertising channel shows you those advertisements is not going to be identical to what is displayed to the user. So sometimes their interface is just blatantly wrong. I can think back to a scenario where I was starting to do some paid advertising for Bluetick through Twitter, and I was using their lead cards. And everything showed up in the Twitter advertisement interface just fine, and I put it out there, and I forget how long it was, but people started sending tweets back to me with screenshots that showed me that they were displaying just a link to the Twitter card, as opposed to the Twitter card itself. And people had to click through it. And at the time I was looking at my stats saying, “Well, these are abysmal, but they’re not costing me very much, so I’ll just let them run.” Because I was still getting email addresses, gnd going back and looking at the screenshots that people were sending me, my ads were just fundamentally broken. So I ended up stopping them at the time just because I was clearly spending a heck of a lot more money, and I didn’t want those images to burn out, and quite frankly it was kind of a distraction at the time. So just that said, you need to be aware that just because your advertisements are tanking doesn’t necessarily indicate that, “Oh, this entire channel is junk.” It could also mean that you’re doing something wrong. So you want to be careful to at least do some spot checks to identify whether or not, “Is it my advertisement that is wrong? Or is it just this channel?” And it’s sometimes difficult to tell the difference.

Rob [25:37]: Yeah, that’s tough. I haven’t seen that happen with AdWords or with Facebook; where the ad showed up different than the preview. But I have to imagine that happens. Especially with Twitter where all the clients are— not all the clients – but the third party clients tend to display stuff differently than the native Twitter things that they own. So I could imagine that there would be differences in display. It’s something you need to watch out for. The next step in how to test is that if you’re going to do this yourself is to educate yourself online. I would just start doing Google searches of “how to run Twitter ads”. And I would try to find like Growth Hackers links from growthhackers.com, or people who are actually doing it, kind of, in your space, because you’re going to find a lot of stuff of people running ads on this network, and they’re doing it for, again, like diet pills, or “make money online”, or just stuff that doesn’t apply to you, and it isn’t necessarily going to be that helpful. But I personally try to buy a course if I can. I try to find a podcast on the topic, and then I try to buy that person’s course – assuming that the podcast is reasonable. And obviously you’ll pay something for that course, but it can save you dozens of hours of trial and error. And maybe it’ll even convince you it’s not a good channel and it’ll save you the time all together.

Mike [26:44]: In terms of courses, there’s a lot of great resources out there that are free. And as Rob indicated you can also go out and look specifically for paid courses from people who are actively working in that particular industry. But there’s a couple of different resources you might want to check out. You can take a look over at Udemy, or Udacity, or at Coursera. And all three of them have courses that are available, either for free or for reasonably low cost, that you can just pop in, learn what it is that you need to learn and get out. And I think the one thing to keep in mind when you’re looking at these courses is that you don’t need to consume every single piece of information that’s there. Skip through, find the things that are relevant to exactly what it is that you’re doing, and then get out and start practicing with that stuff. Because if you spend too much time getting into the weeds, then you can spend forever there and not actually get your ads done. And I think that setting up your ads, and getting the images, and the text, and all that stuff all set up can be very time consuming. And if you’re spending more time learning – and we actually talked about this in a previous episode a little bit; how to balance your time in learning mode versus execution. So if you haven’t listened to that, go back and take a listen to it. I think it was episode 286 where we talked about that. It was “Five guidelines for balancing learning and doing”. So if you think that that’s an issue that you might need to contend with, go take a listen to that episode. But again, you just want to make sure that you’re getting in, getting the information that you need, and getting out as quickly as possible, so that you can start learning how that particular channel applies directly to the product that you’re selling.

Rob [28:09]: Yeah. That’s a good point. There’s a lot more resources on these topics – on ad networks and how to run them – than there were a few years ago. And those course web sites that you mentioned, I’ve actually had good luck finding decent courses on them, and have bought several for my own edification. So the approach I take when I’m testing is that I tend to try to go immediately from an ad to a sale, or a trial, or a demo, to see how that pans out. And then judge is my ad click through rate too low? Or is it the landing page that I’m sending them to that’s too low? Which should I optimize? And typically it’s going to be the ad at first, because your ads aren’t going to be very good. Then I split test images, headlines, and the audience, until I get to the point where I’m paying a reasonably small amount and getting a decent amount of traffic. By that time you’re going to find out that your landing page sucks. And then I’m going to use Optimizely to improve that. If I then spend several weeks trying to improve the landing page, and trying to get them to buy or to try, and that’s still not working, then my guess is that either the traffic is not qualified enough, or I’m just asking for the sale too quickly. So then I shift into this “Plan B mode”, where instead of asking for a sale or trial I ask for email opt-ins – either to a webinar or to an email course. Then nurture those leads, see how many convert, and this is where a tool like Kissmetrics or a Drip – because that’s what we use it for – where you have tags, and you can actually report on everyone who’s coming through those ads – because their tagged as their source is that ad. Then you can see, “Boy! Did they sign up for a trial? How many of them did? Did they sign up as customers? How many of them did? How much have they paid us?” It’s very, very big, because it’s hard to do that without a tool where you can actually see individuals and not just aggregate data. That’s a quick run through; the philosophy of actually running a test is to first try to ask for the sale trial, figure out what’s not working, fix the ad, then fix the landing page, and then – if you get that working – then it’s awesome and you can scale it up. And if that doesn’t work, then you switch into this plan B mode of getting the email opt-in and nurturing. So it sets you sale cycle out longer, it means there’s more steps between them initially clicking the ads. And, not only steps, but more time in them signing up and buying from you. But that’s what this is about. It’s about finding and reducing friction during this process.

Mike [30:23]: And I think something that that process kind of highlights is the fact that you’re going to have to go through this several times in order to find out whether or not a given channel even works at all. Because one of the first things we talked about was making sure that you had enough budget to go through and run these tests. And when you’re testing them, make sure that you’re taking a look at the statistics, and identifying your conversion rates, and how much you’re paying for the incoming leads, and whether or not those people convert to trial. And it’s time consuming to gather that information, and analyze it, and figure out where the leaks are in that particular sales funnel. Again, it’s just a matter of going through, gathering as much of that data as you possibly can, and then iterating through it multiple times in order to find where those leak points are. Because you’re not always going to find them the first time. Sometimes it’s misleading it best. I mean, you’re going to look at those things and say, “Well, this particular channel isn’t working.” And I think that that’s where a lot of the fear, uncertainty, and doubt comes about from some of these different ad channels. Because somebody will try them out and say, “Oh, well, I tried that and it didn’t work.” And your response to that should be, “Well, what did you exactly do? And why did it not work out?” And if somebody can’t explain why something didn’t work then they may not understand all the subtle nuances about it, or they just didn’t dedicate the time to it. A lot of these ad channels will work if you spend the time and effort to optimize them. Now that said, there are some that are not going to based on the type of product that you have. But if other people are advertising on those ad platforms, chances are good that it’s working for some of them, because if it didn’t work for anyone then nobody would do it. So nobody’s going to spend money to make nothing. That’s obviously a losing strategy.

Rob [31:57]: I think a final thought is that if this sounds complicated and scary, it’s not. But it does take work and it does take time. And just the dream of plunking an ad on to AdWords and paying five or ten cents a click, and then just having this amazing cost per acquisition is not—I’ve never seen that happen. It does take work, and it does take refreshing of the ads. So if you have the time to do it, invest, but consider it another marketing channel. It’s not anything that’s going to be easier than other marketing channels, or particularly less time. The nice part about it is if you get it to work then it tends to scale really well, and you can drive a lot of trial pretty quickly. I think the last thing I’ll throw in is if you’re not already doing retargeting, that is actually where I would start before I got into all this stuff that we’ve said here. I would consider using a tool like Perfect Audience, or going directly to Google AdWords, or to Facebook, and they have retargeting images build into them where you can get a pixel, and then essentially cookie people, and then retarget them with ads as they travel around the Internet or on Facebook. Those are going to be your cheapest clicks and your highest converting clicks. And so that’s actually pretty easy, and there’s not too much magic to it; maybe some testing of the ads and such. But you don’t have to get into all the targeting, which is nice, because you just have this audience. And that would be, I think, the low hanging fruit in terms of paid acquisition. But if you’re trying to get new visitors, that’s really more of what this episode was focused on. Because to do retargeting you need to have a nice bulk of existing visitors already. You can’t just start from zero, because there’s no one to retarget.

Mike [33:25]: I think that about wraps this up. If you have a question for us you can call it in to our voicemail number at 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 a full transcript of each episode. Thanks for listening, and we’ll see you next time.

Twitter Digg Delicious Stumbleupon Technorati Facebook Email

Comments are closed.