Episode 83 | HitTail & AuditShark: Integration Marketing, Logo Design, Survey Analysis and Leaky Funnels

Show Notes



[00:00] Rob: This is Startups For The Rest of Us: Episode 83.

[00:04] [music]

[00:12] Rob: Welcome to Startups For The Rest of Us, the podcast that helps developers, designers and entrepreneurs be awesome at launching software products, whether you’ve built your first product or you’re just thinking about it. I’m Rob.

[00:22] Mike: And I’m Mike.

[00:23] Rob: And we’re here to share our experiences to help you avoid the same mistakes we’ve made. What is the word this week, sir?

[00:29] Mike: I am undertaking a lot of work to create work for others right now.

[00:32] Rob: Yeah. [Laughter] Yeah, this is the work that you can’t outsource, isn’t it?

[00:36] Mike: Yes.

[00:36] Rob: It’s outsourcing your work you can’t outsource outsourcing your work.

[00:39] Mike: Yup.

[00:40] Rob: So what have you been up to?

[00:41] Mike: So I’ve been going sort of a hiring binge lately. I hired somebody to build out the AuditShark sales website for me. And essentially just come up with the design for it. I’d say for the last year or so, I basically have the same sales website out there. I really haven’t spent a lot of time or effort on it. Right now, I’m kind of focusing my efforts on building something that actually looks good and addresses the paying points that have been kind of brought to my attention that people are willing to pay for from that Wufoo survey that I ran. So I hired somebody for that and right now I have a job opening for another .NET developer to add to the person who I already have working full time on the AuditShark code that he’s working on. So hopefully, by the end of this week, I’ll have two people working on that and the other person will be working on the website. And then I’ll have to start generating the content and stuff that’s going to go in to the framework that he’s building for the website.

[01:34] Rob: You have a WordPress site up right now, right? Just a WordPress theme that you customized like a year ago or something —

[01:39] Mike: Yeah.

[01:39] Rob: And you had the logo design, you know, by someone. So what’s the rationale at this point to go back to the drawing board? Do you have like a much better design for it or I guess why would you spend a time now to go back and do that rather than use what you have?

[01:53] Mike: There’s a few different reasons. One is to kind of switch it off of WordPress and I’ve started integrating KISSmetrics to kind of all throughout my website. And one of the things that is — I don’t know. I won’t say it’s a real pain in the neck to do it with WordPress but I don’t like the — the look and feel that comes out. I’ve seen the — the particular theme that I’m using quite a bit out there and it’s starting to wear on me a little bit. So I’d like to have a little bit more control over it. And the other thing I’d like to do is I’d like to make the sales site and the application itself look a little bit more like they were built together because right now, they just don’t. But you know, it’s — it’s something that I’m going to have to do eventually and if I get started now, then it’ll  be ready in like the next couple of months because it’s not like you can just say, oh well I’m going to redesign this and a couple of days it’s done. I mean it takes time to get it done.

[02:44] Rob: Right.

[02:44] Mike: So I’m spending the time now to find somebody to do it and then, you know, hopefully, in the next couple of months that it’ll — that it’ll get done and I won’t have to worry about it.

[02:53] Rob: Got it. And in the meantime, are you continuing — you had some survey results and you had fifteen or twenty e-mails I think and you were going to contact those folks and kind of do some customer development interviews to find out more in depth about maybe their needs and how you would structure your copy on your marketing site as well as potentially what you would build in to the product. Have you continue with that?

[03:16] Mike: Yeah. So I started talking to people and getting a little bit more of an understanding of specifically what they’re looking for. And one of the things that I’ve come across is that in both the copy that’s on the website and in my discussions with people, it’s not inherently obvious exactly what the product does and why it’s important. So one of the things that I came across was that the fact that AuditShark will dynamically look at your machine and figure out what it’s supposed to be looking at is now something that’s clear.

[03:45] So for example if you have a patchy installed in your website, then it will run all these different things on your machine to make sure that the patch is lock down and you know, if it’s a Unix or a Linux machines, it will track different places to find out what user accounts had been created and then we’ll monitor those overtime to make sure that they haven’t changed because one of the things that people should be looking at in order to secure those systems is our new user accounts being created. And if so, you know, what are those user accounts? It needs to notify you and those are the things that you should regularly do at some point anyway but most people don’t. But if a new user popped up on your machines, obviously, that’s something you would probably want to know about. If you did it, that’s fine but if you didn’t do it and you just get this notification that says, oh, there’s a new user account that’s been created and has got, you know, UID of zero, well, that’s probably a problem.

[04:36] Rob: Got it. Hey, I have a question. How many hours do you estimate that you’re going to have to spend to basically kind of repurpose the code that you have the AuditShark code or expand it to fit the new product idea that you’re exploring? Or do you know that yet? You know, the full feature set.

[04:50] Mike: I don’t specifically know that yet. I mean a lot of the stuff is going to stay the same and I haven’t even really look at what needs to fundamentally change in order to meet that new model. I don’t believe it’s a lot but not a lot is not quantitative I’ll say. I would say maybe fifty to a hundred hours but it’s probably fifty to a hundred hours that I mostly likely would have spend anyway because there were a lot of things that just weren’t done. So if the things would work and they would do what they’re supposed to do but there are use case scenarios that I hadn’t accounted for yet and this particular one is one of those — use case scenarios that I had thought of them like, hey, I think the people would want to use this for, you know, their web servers and I didn’t have a use case scenario for them being able to install at on a web server. It was solely around them installing it and like an active domain environment and then running there.

[05:42] So this is a use case scenario I had always scoped out and thought of ahead of time and I don’t look at it as being extra work that I wouldn’t have done had I gone in this direction. I think I would have done it anyway. It’s just the matter of timeline I guess.

[05:54] Rob: Right, that makes sense. Have you thought about just putting up a landing page? While you’re doing these interviews, you know, you do have a decent idea what people might want but you could get a pretty attractive landing page that’s not use by anyone, you know, like off of themeforce.net. They have, I don’t know, like seven to twelve bucks in there. There’s some pretty nice one of these days. You know, John Turner has — he’s an Academy member and he’s doing really well selling his ultimate coming soon plug-in which is cool because  you can then have a theme on the back end and then just have this plug-in that’s, you know, swaps at a landing page as your home page. But I guess you’re moving away from WordPress but for now, you know, you could keep the WordPress install and use his plug-in and just have a few senses describing the specific paying points.

[06:34] Mike: Uh huh.

[06:34] Rob: And then push off the website construction until — the marketing site construction until after you’ve had more time to talk to these customers and potentially have even gotten started on updating the code with the final feature set and then kind of set someone off on — on building it. Have you consider that route?

[06:50] Mike: Not specifically, I mean one of the things that I was a little bit concerned about in doing something like that is the SEO that I’ve done on my existing web page but I don’t know is that’s a huge hurdle to overcome. The bigger issue is kind of putting things on to that coming soon page that makes sense in — and or in line with what people are looking for. You know, some of that I think I’m still kind of straightening it out but I think that there’s probably enough interest in what I’ve talked about so far. I mean virtually every single person who answered the survey with I think two exceptions out of about thirty that said, “Yeah, add me to your mailing list.” That side of it I think turns out pretty well. But I don’t have a good reason for not doing that so —

[07:30] Rob: Yeah, I mean —

[07:30] Mike: … that’s probably something I move forward with.

[07:32] Rob: I see two kinds of benefits to doing that over going at in building the full pledge site right now.

[07:37] Mike: Just a correction though, I’m —

[07:38] Rob: Yup.

[07:38] Mike: … I’m not having the whole site built up. I’m having basically a template build for me with all these different HTML and CSS Snippets for different things that I may want to put in to the site —

[07:48] Rob: Yup.

[07:48] Mike: … and those will go in to like a full blown .NET web application that I’ll kind of put together. Why I put the contractor on is just basically building out the design for it but not the actual content.

[07:58] Rob: So I guess that I want to see the benefits in the landing page. Does it focus you’re — you’re just going to get more e-mails from it, period. You know, any visitor you send there is going to read it, read the few bullet points you have and then either give you their e-mail or not or is it if build out more of a full site after navigate through to give you their e-mails. So you have a higher conversion right there. In addition, I think — I think you’re right. You’re still kind of an early in the journey of figuring how exactly what AuditShark is going to become. And so not too hard to put together some bullet points a few senses to describe again. Probably harder to build out, you know, the full three, four, five page site with the tour and all that stuff.

[08:31] Mike: Right. That’s a good suggestion. I think I’ll do that. What about you? What have you been up to?

[08:31] Rob: About a week and a half or two weeks, I mentioned that I had rolled out Chinese character support and actually Chinese Search Engine Support which is the two different things but they’re related.

[08:45] Mike: Uh huh.

[08:45] Rob: I rolled them out with HitTail. So now, like these search engines like Baidu and this one called the Sogou and just a bunch that I can’t pronounce frankly, they now worked with HitTail which is cool and I have an “enterprise client”. My first four figure a month client trying it out now.

[09:02] Mike: Cool.

[09:02] Rob: So we’ll see how that appends out. I’m like cautiously optimistic. I tend to be skeptical when like really good things happening or like a “big break” but it could be — it could be a nice one, you know, it can be good for revenue. So he’s about seven days in to a trial. I just touched base with him so…

[09:18] Mike: Cool.

[09:18] Rob: That’s kind of the update on that. Now what I’ve really been working on for the past seven to ten days is this one-click article support because I mentioned last time that one of the main reasons that people were — were canceling their HitTail accounts was because it was adding stuff to their To-Do list, right? It gives them the suggestions that they should target but they don’t want them to go out and either try to write themselves or hire a writer. And so there are a few services that basically have these big pools of writers and you know, they offered to write articles for you and there are a few of them with API’s.

[09:51] Mike: Oh really?

[09:51] Rob: And so — yeah. So integrated with — with one of them that has 80,000 writers, not all are active but they have huge pool of writers. And so, you know, it’s kind of a no brainer for someone like me to — to integrate there. There are some other services, a lot of by the services that integrate with them and kind of use them as their back end Mechanical Turk almost, right? It’s like the turnaround is one to two days depending on the length and get a full pledge article and there’s like a bunch of different pricing tiers and all that stuff. But I made it super simple. I’m calling this an MVF. It’s a Minimum Viable Feature.

[10:21] Mike: [Laughter]

[10:21] Rob: Instead of actually — and when I sketched to that, of course, you know, I’m the developer entrepreneur. I was like “Oh man.” I want there to be all these choices and you could do 300 to 400, 500, 600-word. You can say I want SEO or not. How many times should it, you know, should these keywords appear in there. I want the people to be able to submit it and then review it and preview it and accept it or reject it and that — you know, there’s always options available to their API and I kind of cranked up at estimate and realized there’s no way I want to build all of that. So I just — I chucked it all. I said, “When you click this button, we charge your card on file 18 bucks for one 400-word SEO article and it comes back in two days.” And there are a few options, you know, you can change the title and change the keyword and do a few other things but it is super, super stripped down and if this works and it appends out and A, it helps retention and B, it actually, you know, generates the revenue because obviously I’m marking up the service a bit so there’s a little bit of margin for me.

[11:15] If that works, then I will go on and expand it and certainly add it, then adding it, you know, and making it a 300 and 500-word options. It’s not actually that hard at this point. I mean I kept all that stuff in mind while I was coding but I just — to do all UI work to support that would also really make the form a lot larger. So I pushed that live yesterday and it, you know, I’m super stoked that that’s finally done. It’s just — it’s kind of the biggest new feature that I built since I acquire the HitTail.

[11:39] Mike: Uh huh.

[11:40] Rob: And I haven’t launched it like I haven’t announce it to anyone. It’s soft launched now. I had placed one order myself just to make sure everything works. And so far that’s good. And then I think I’ll slowly start e-mailing customers in like ten or twenty person chunks and just kind of see how it goes before I really do a big blast in the promotional campaign.

[11:58] Mike: Uh huh. Now, you’re going to be measuring how many people kind of go through that funnel whereas you start telling more and more people about it.

[12:05] Rob: Yeah like how many people see the page —

[12:07] Mike: Yeah.

[12:08] Rob: … and click versus not. Yeah, I will be. Yup. So I have — I have KISSmetrics to install. I also have Google Analytics. But you know what? Good point I should actually install maybe Crazy Egg or even Inspectlet. Inspectlet is like the inexpensive version of ClickTale where does a screen recording of people doing stuff.

[12:23] Mike: Uh huh.

[12:24] Rob: And Inspectlet supports JavaScript and divs and all that because I have — it’s basically a Lightbox that pops up. So I probably should do that just to watch people on that specific page. I may —

[12:34] Mike: Yeah because if they’re not seeing it, then they’re not going to, obviously, click through it but you want to make it visible without being I’ll say intrusive.

[12:40] Rob: Yeah. And I’m planning on adding like a little new, you know, icon at the top in HitTail, you know. There are grids that you can sort and you can page and do that stuff but there’s not a lot of adornment on purpose because we want it to be simple and fast but if I put a big new icon at the top, I’m hoping it’ll draw to people’s eyes.

[12:58] Mike: Cool. Hey, did I tell you that I’ve did a contest on 99designs for new AuditShark logo?

[13:03] Rob: You did. You mentioned that to me.

[13:05] Mike: I finally went through and actually approved one and got all the files and everything transferred. So I’ve got that. I already sent that off to the designer who’s working on the sales website template that I’m going to using and hopefully, he’ll be integrating that in shortly. And I’m hoping to get some screen shots and stuff back just to kind of see what he’s working on in the next couple of days. But the logo I think turned out really, really well. I ended up with I think 110 different designs to choose from. There were probably three that I thought were in my top three. [Laughter] It was basically those three that I kind of work through a couple of different iterations and there was a fourth one that I kind of threw in at the end just making in to the finals.

[13:46] If you’re not familiar with how 99designs works when you’re doing a logo contest, basically, you pay a specified amount whether it’s a bronze, silver or gold contest and you pay more for the gold contest than you do for the bronze. And the one I did was I think $500 and the people who just start submitting their designs to you and you get to rate them and comment on them and send them back for revisions or whatever and they’ll just keep submitting them to you as long as you are I’d say pretty diligent about providing feedback about what you like, what you don’t like when it gets to a certain point, I think that’s about three and a half days in, they kind of draw the line. You have to pick your designers, who you want to move on to the finals.

[14:24] There are certain designs that people would come up with and you just don’t like anything that they come up with. So — those are the people you wouldn’t want to bring on to the finalist stage and then of those, I just picked four to go on to the finalist. And then of those four, I picked one of their designs and I had them tweaked a couple of things and what was it — I think five or six days and it was done.

[14:41] Rob: How cool. How much to that one have cost in?

[14:43] Mike: It was 499. So —

[14:45] Rob: Cool.

[14:46] Mike: … $500. One of the things that was kind of interesting was I put it on Twitter that I had been drawn in this contest and somebody called me out on it over episode 71 via Twitter. And if you will recall episode 71 it was “Things you shouldn’t pay for early on in your business.”

[15:02] Rob: [Laughter]

[15:03] Mike: You have that?

[15:03] Rob: You know, it’s funny. I was just going to mention this because that’s one thing I don’t pay for. I don’t have logos for anything, like I do it on purpose now. Like the Academy has no logo. The only reason HitTail had a logo is because the designer who’s designing had said, “I need to put a logo here.” And I said, “Just put a text.” And he said, “No, I have to do it.” So he just did one off logo. But yeah — so I was going to ask you about that and give you a little crap about it but I’m glad someone else [Laughter] headed me off.

[15:27] Mike: No, I mean he did and — you know, it’s a good point. I think a lot of the things that I said, there are still a whole true. I mean which was primarily if it’s not making you money and it’s not going to make you money, then what difference does it really make? But at the — at the other end of the spectrum is, you know, I would still have to have something there and I would have to have some. The very least stylize text and you can look at stylize text and say, “Well, that’s a logo.” And as like, “Well, I don’t really want that.” It was more of a personal preference on this particular product. In looking what I ended up choosing, I think that it fits really, really well.

[15:59] Rob: And here’s the thing, you can make general rules like we talked about 71, you shouldn’t spend money on stuff, you know, before you need to spend it. But the bottom line is there are a couple of things, one, as an entrepreneur, you can have some indulgences. If you really wanted a nice logo, then go buy yourself a nice logo, you know. But when you’re doing it, you know that it’s not going to dramatically increase your click-through rates to your signups or anything.

[16:19] Mike: Right.

[16:19] Rob: Just know that going in to it that you spent that 500 bucks and it’s not necessarily going to have a high ROI but if you needed that to like feel excited about the project, then do that. That’s not a big deal as long as you know it.

[16:28] Mike: Well the other thing that does — as you said, I mean if it excites you about the project to actually do that and it really did. I mean seeing the logo kind of help energized me a little bit and put me in the right frame of mind to kind of go on and tackle a lot of other things that I kind of had on my plate for AuditShark. So I think it was helpful for me but at the same time one of the things that I think is important about AuditShark is because it is kind of a security product, I think that it lens credibility to it by having a nice logo. In many cases, a logo is probably not necessarily going to land you any sales but at the same time I think for this particular product, having a logo is going to help it land sales that might have otherwise been on the fence.

[17:06] Rob: Right. You can definitely take it too far and you know, I talked about indulgences and doing stuff just to get you excited and you can spend a hundred of hours or a thousand of dollars doing something stupid just because it gets you excited. And obviously you don’t want to go that far. But you’re not doing that here. And I would agree with you like we’ve talked about different ways kind of a tiers of building sales websites and at the very bottom by an HTML template right above that you can buy a premium WordPress theme. Above that you can get a — a kind of a low end custom-design and you can go a higher end and we’ve done — I’ve definitely done all of those tiers and I know that they impact click-through rates and they impact conversion rates but for some of them like when I read at HitTail, I paid top dollar.

[17:45] I paid the most that I ever paid for a design ever, [Laughter] you know, in all of the sites that I built. And that was a little bit of an indulgence but it was also a little bit of me getting really excited about the idea and wanting something that I could be proud of. And third, it was like you said, it was about building some credibility for this thing because it is kind of an — it bring you something to another level, right? It’s not some really tiny niche product. It’s something that you do want to have a lot of credibility and therefore, there are some — there’s some give and take there. There’s some things that may be on the boarder of, you know, how early you should have them done and how much you should pay for them and if you have the money and you really are going to — kind of go bigger and go with the bigger idea, then some of those things can be done earlier on or you can invest more money than you otherwise would have if you might be going after, you know, more of a niche idea or you’re a more cash strapped.

[18:30] Mike: Because I know that there’s money here, I think it makes it a little difference. So if this was a new idea that I didn’t know if it was going to fly, that’s when I would probably be a lot less likely in a way to spend the money on, you know, the logo. In just drawn in the survey, I’ve got anywhere from 1500 to $2500 worth of commitment so far based on, you know, the survey. I can definitely see that there’s money there. So it doesn’t bother me to spend $500 on making it looked nice.

[18:58] Rob: Right. Now, you said something that you said because I know this is going to work and I think, I personally think that’s taking it too far at this point.

[19:05] Mike: Uh huh.

[19:05] Rob: I do feel like you’ve gotten some positive affirmation of your hypothesis —

[19:09] Mike: Uh huh.

[19:09] Rob: Like to me, at least maybe I’m a pessimist or something, in the back of my mind it’s like, no, you still need to prove this out.

[19:14] Mike: I guess no, it’s probably not necessarily 100%. It’s I’m fairly confident that I’m on a right track. Not necessarily that this is going to work and I know exactly what I’m doing. It’s more of that I’m on the right path. That’s more what it is I think.

[19:26] Rob: You’re encouraged, yeah.

[19:28] Mike: Yeah.

[19:28] Rob: And that I would agree with.

[19:29] Mike: Uh huh.

[19:30] Rob: So I went in to a little indulgence of my own last week. I was knee-deep in code doing that one-click article feature and just for kicks, I wanted to try a little bit of paid customer acquisition so I set up some Facebook ads and had kind of a blast with it. I forgotten how much work it is to set up and maintain Facebook ads because they just burn through people so quickly.

[19:53] Mike: Wow.

[19:54] Rob: For those who have — yeah, for those who haven’t —

[19:56] Mike: I like the — stock price as it goes down, right? [Laughter]

[19:58] Rob: Oh my gosh. Tell me that’s a disaster. [Laughter] Anyways, that’s a whole other podcast but —

[20:03] Mike: Yup.

[20:04] Rob: … but it was fun. With Facebook ads, you target a demographic and you tried to get the demographic to a certain size and then you really got to go for the visual — the visual and the headline or — for the most part, all accounts to get the — the attention. But then you basically, a Facebook just shows the ads over and over to the same group. If you niche it down properly, then your number of people that you’re displaying it to is not huge. And so Facebook will repeatedly show it to those people. And so if your click-through rate goes high, your cost per click goes low but then pretty soon, your click-through rate just plummets because people get blind to the ad because they’ve seen it so much.

[20:36] So realistically, like if you put an ad up and you have, you know, an audience size based on the demographic you’ve chosen of around a hundred thousand people, it’s like two or three days and the ad is done. And so, I put up three or four ads, A/B testing a bunch of different stuff and they were just gone in like two and half days. And then I put up another — I put up like nine the next time. And I had a bunch of, you know, image I was — images. I was using different headline combinations and testing and it was fun.

[21:01] I was spending about twenty to thirty minutes a day on it and in total, I’ve dropped — I guess it got pretty good click-through rates about 50 cents a click and I spent about 300 bucks. Yeah, it was a fun experiment. I wound up getting a bunch of customer, well, a bunch of trials, anyways. And that my current trial to paid conversion rate it looks like I will be about — it was about six months of a cost of the average revenue to acquire a customer. So if that makes sense so I —

[21:28] Mike: Yeah.

[21:28] Rob: … it’s about six months of my average revenue number in order to acquire each of those customers assuming they convert the way my other — my other customers do from trial to paid. So now, I want to get that down to below that. It should — I wanted to be more around three. Since I’m bootstrapping, I don’t have this huge pool of cash to dump on it but —

[21:44] Mike: Uh huh.

[21:45] Rob: … it was a fun experiment. I shall of — it was way too much work. The thing was is for the amount of work and even, you know, kind of the money I was investing there are so many other options at this point that would work better than that.

[21:58] Mike: I can see that. So when you’re I guess doing this Facebook ads, I mean do you have a lot of control over how much I guess gets spent or how much they get showed to people? I guess how much control over the process do you have?

[22:11] Rob: So first, you specify all types of demographic information, right?

[22:15] Mike: Uh huh.

[22:15] Rob: What they’re interested in and what they have in their profile and their age range, their gender, you know, all these different stuff. And then you can do cost per click and then Facebook will just show it as much as they think — they have an algorithm obviously that shows it and if your click-through rate is high right off the bat, then it’ll drop your cost per click and they’ll show it a lot more because they want to make — it makes them money and it makes you money.

[22:38] Mike: Uh huh.

[22:38] Rob: And then you can also choose cost per thousand, right, CPM —

[22:42] Mike: Uh huh.

[22:42] Rob: … where it just — it’s not cost per click but it’s actually cost per thousand impressions. And that way with a little different, they still had an algorithm but they seem to show it a lot more. They kind of didn’t care about the click-through rate, right? Because that’s not they’re billing on and so that was — allow me to get to add in front of more people and so I tolled around with both of them. I got on this one, I got the better click-through rate using the cost per click and in the other campaigns I’ve done, I’ve had CPM worked better. So it varies from time to time. I’m sure there’s — as a rule, certain people can get a phenomenal click-through rates or CPM I’ve heard and I’ve seen that happened on one or two campaigns. And then from there, you set your daily budget. That’s how you — how much you show, right? Once it’s out of money, then it stops showing.

[23:25] Mike: Got it. Yeah, I was just kind of curious how Facebook’s model for advertising stuck up against like Google AdWords. So…

[23:32] Rob: Yeah, it’s definitely getting more sophisticated. It was pretty simple about a year or a year and half ago but they’re adding more features and I think the algorithm is getting a little better. I mean I imagine now they have IPO’ed, you know, they really need to kind of take this up a notch.

[23:45] Mike: Yeah. And I would imagine that they probably have a lot more data about the users that they’re showing this too to figure out whether or not that’s going to fly for them or not. So…

[23:45] Rob: Yeah, it’s so with me.

[23:54] Mike: Yeah, I’ve been doing some data analysis on my end for the Wufoo surveys that I’ve did for AuditShark and one of the things that I found that it’s interesting but not surprising is that the people who have Window servers are willing to pay roughly twice what the Linux admins are for the same thing.

[24:12] Rob: Interesting. Yeah, that makes sense, right?

[24:14] Mike: It makes sense but it also makes pricing difficult, you know?

[24:17] Rob: Yeah.

[24:17] Mike: I mean I had roughly the same amount of people who had Window servers as had Linux or UNIX servers. So it’s —

[24:17] Rob: I was going to ask you that.

[24:26] Mike: Yeah.

[24:26] Rob: But that’s some — you know, I was talking, I mean this was a couple of years ago but Joel Spolsky had said — we’re at the Business of Software Speaker dinner. He and I were sitting two chairs down from each other and there was this whole discussion about Linux versus Windows users and all that stuff.

[24:40] Mike: Yup.

[24:41] Rob: He basically said that their Linux users of FogBugz were only — it was a very small percentage. It were somewhere around 15% or 20% but the day, at least for them for their market and actually accounted for a substantially higher percentage of their support issues. They must have — some people had said, well it’s because they are more hobbyists, you know, they’re in to PHP or whatever so they’re not going to be in to the Microsoft stuck looking for a cheaper solution and they don’t or maybe not as enterprise. You don’t know what they’re doing as much. Other people had said, well maybe your Linux product isn’t as good or it crashes more something and he, you know, contented that it didn’t. So that’s an interesting thing to think about too. I wonder if there will be any kind of ease of views or higher support burdens if you go Linux versus Windows or go both.

[25:27] Mike: Yeah, that’s one of those things that — it worries me a little bit just because if — if I kind of go down on the price in order to gain those Linux customers is that they’re going to kill me on support side of things.

[25:38] Rob: Yeah, seriously. And so you’re not going to charge 99 bucks for Linux and 199 for Microsoft Window?

[25:44] Mike: [Laughter]

[25:46] Rob: I would have to lean towards going Windows only when you launch then.

[25:49] Mike: Yeah.

[25:49] Rob: I got to go for the higher end to start with, right?

[25:52] Mike: Right, right.

[25:53] Rob: Especially if you have the equal amount. I was going to say if you have more or the same amount of Windows.

[25:57] Mike: Yeah. I think I have to go back and double check the spreadsheet but I think that they were — it was roughly equivalent in terms of the respondents. And I know that there was one person who had come in and said, “Yeah, I pay for it” but it really skewed the numbers because they had like, you know, 75 Linux servers or something like that. So that made up the bulk of them. But in terms of the actual number of people who chose them, it was significantly less. But yeah, I mean the distribution I am looking at it now. There were eight who had said Windows on this spreadsheet. So on this particular one, there’s eight of them in here that say Windows and then sixteen who said Linux, you know –

[26:36] Rob: That’s not 50-50. [Laughter]

[26:38] Mike: It’s not but the people who had Window servers tended to have more. So for the couple of exceptions there was the one that like I said have 75 that skewed at pretty far on the direction of Linux but otherwise, it was — it was mostly the — and the other thing is people going to answer more than one thing. So there were several people who had both Windows and Linux.

[26:56] Rob: What are you going to do? Do you think you’re going to come out Windows only and do the higher price to start with?

[27:00] Mike: I’m still trying to figure it out because one of the things I’m looking at is a lot of the competitive products that are out there I mean because I asked what are you currently use for monitoring on your server because Ted from Moraware had I kind of ran my survey past him first and just to get some feedback on it and he’s like, “This looks good.” The one thing I’d asked is what other monitoring solutions people pay for already because that might be kind of a self selecting crowd of people who are willing to pay for those types of things which —

[27:26] Rob: Sure.

[27:27] Mike: … gave me a great idea. I was like, oh, well if you’re already paying for those things then, you know, it would makes sense for me to either integrate with them or partner with them in ways that would allow me, you know, use my software alongside of theirs, et cetera. People pointed out what other tools that they use to monitor those servers. So I got a lot of information that I probably wouldn’t have gained otherwise and you know, just been looking at some of those offering or what they do. And some of them I just never heard of before and then other ones like, you know, Pingdom and Cloudwash . I mean those things, you know, pretty straightforward and I call them more well-known. You know, it’s just interesting looking at all that data.

[28:02] Rob: Yeah, it is. And you know what that helps you is this concept that I’ve been telling in one of my talks called Integration Marketing and there’s a few entrepreneurs doing it really well right now and I’m actually diving in to that my next marketing/technology challenge with HitTail is I have four different companies that want to integrate with me so — that I will either create an API and have them ping in and grab some keyword suggestions or you know, go out and integrate with their app market place.

[28:31] And the cool part of it is that I’m only integrating with people who already have customer basis and are willing to either e-mail or blog about the integration. And so I get this link and a mention and it sends a bunch of traffic, you know, it’s not — it doesn’t have a long fly wheel effect but you know, it — what it brings is this nice big chunk of people who kind of signs up for the — for the product and since it’s a SAS app, you get the recurring revenue from them. It’s not I mean really the cause is to do the integration, you know. So either I pay someone to write the code or I write the code myself and that’s where you in your survey have just asked about future integration partners, right?

[29:07] Mike: Uh huh.

[29:07] Rob: And if they’re already using Pingdom and these other things, whether you actually do a physical code integration with them or you just basically just to have these customers and I’ll them know about you and you have your customers and kind of do a joint venture e-mail type of thing. That’s a great way to go. It’s kind of a low cost guerilla marketing tactic but it works out really well. That’s the next steps for me after I get this whole article thing launched and I’m going to do a little promotional over the Chinese character thing as well and then move on to these things which had been on the book since like February. I started getting contact to them in November-December from a few shopping cart vendors and a couple other companies that will remain nameless right now but I’m pretty excited about — about doing that. These companies have few of intensive thousands of active paying customers. So —

[29:52] Mike: Uh huh.

[29:53] Rob: … could be — could be a nice — nice expansion of the customer-base.

[29:56] Mike: Yeah, definitely building an API that people can kind of interface with and [0:30:00] get data in or out of your product is helpful in leveraging your product is more of a platform, you know, as more people do it, it’s just gives you a larger breath, increasing your luxe surface area I believe.

[30:13] Rob: Right, from texting. But you know, in terms of the API I would never go out and just build an API and say, “Hey, here’s API. People can use it” because that’s an — that requires marketing too, right? You have to go out and recruit people to do it.

[30:25] Mike: Right.

[30:25] Rob: Only reason I’m building one is because I have four people onboard ready to do it now and I vetted them. They have, you know, between 500 and tens of thousands of paying customers and they are ideal integration partners because they are like shopping cart vendors who are great candidate for HitTail because e-commerce sites get a lot of value out of it and then there’s a few others that are marketing type software that were just the keyword suggestions really, you know, would work well and a couple of them, again, I am integrating in to a very consistent and a couple of them are going to pull for me and need an API from my end. So but yeah, I would never just go out and build — you do see this kind — it was like the web 2.0 thing, right, was to become a platform and build an API and I don’t think that’s a good move straight upfront. I mean I think you have to — you have someone to need it before you build it.

[31:09] Mike: And those people who need it, you simply work with them to figure out what it is that they need and you build —

[31:13] Rob: Exactly.

[31:13] Mike: … from that stuff in to an API and then as other people start using it and leveraging it, then you build out other things on top of it. You don’t —

[31:21] Rob: Exactly. Dude, that’s right. And my API right now it’s all built. It’s one call, one method call, get suggestions.

[31:28] Mike: [Laughter]

[31:29] Rob: You know what? There’s authentication and then they’ll get suggestions because that’s all I need right now. And now and some days someone will lead some dynamic tracking code and then — then we’ll build that out so it’s an MVF as well, Minimum Viable Feature. I think I need to write a blog first about MVF because I Googled for it. I’m not — someone surely has talked about this before and of course, it’s nowhere to be found.

[31:47] Mike: [Laughter]

[31:48] Rob: Maybe we’ll rank for it with the transcript to this podcast.

[31:51] Mike: Yeah, maybe. It’s funny that you mentioned building an API because with AuditShark a way a lot of the data is pass back and forth that’s all done through web services. So in a way, I almost have an API that’s kind of sit in there. It’s just not documented for anybody else to use.

[32:08] Rob: And you need the authentication later on their security authorization authentication.

[32:13] Mike: You do but I use a token for just about everything.

[32:15] Rob: Oh nice.

[32:16] Mike: And that token is built in to the product. I think it can be pretty easy to surface that in other locations but I think mine is probably different than most API tokens and that my token is only valid for about an hour. So —

[32:29] Rob: All right.

[32:30] Mike: … most — most services you signup for, they have this API token and you know, you can click a button and reset it at any time but otherwise, it’s valid until the end of time. My system when you pass things back and forth establishes a token which is randomly generated for you and that token is valid for the next 60 minutes. And if it times out while you’re in a middle of doing something, then the next function call will basically say, “Hey, this token has timed out, you’re going to have to request a new one.”

[32:55] Rob: Nice.

[32:56] Mike: So —

[32:56] Rob: Yes, it sounds like you have something that built in already.

[32:58] Mike: Yeah.

[32:59] Rob: Yeah. Have I talk about operation retention?

[33:01] Mike: No, what is it?

[33:02] Rob: This was like the big movement. In January when I relaunched HitTail, I got this big influx of traffic and then thirty days later when the trials ended, I got to start seeing the conversion rate from trial to paid and then I started watching the churn rate of paid customers how many canceled per month. And right away, it was not good. I was just clutching my eyes as I watched [Laughter] these numbers go in the wrong direction and they were just bad. They weren’t anywhere near the goals that I had. So I’ve started doing a bunch of marketing and a bunch of people coming in but I have this leaky funnel, right? And it was better than the old HitTail design. It was better than the old HitTail trial, all that stuff but it wasn’t where I wanted it to be.

[33:41] So I sat down and wrote in my notebook. I said “Operation Retention.” All I’m going to focus on, I’m going to stop all marketing. I’m going to stop anything that takes my time and I’m going to focus on getting the churn rate down, driving the trial to paid conversion rate up because the number of visits to trials was great like that’s — it’s totally in a rate of where it should be. I could tweak and get a little more out of it, you know, just test copy and A/B and stuff but their big glaring bottlenecks the big holes where the trial to paid conversions and the cancelation rate early on.

[34:14] So I wrote down eight things. I contacted customers. I had my VA contact everybody who canceled to ask them why they canceled. I put together a spreadsheet and I figured out that there were seven or eight main reasons people were objecting and some of them included things like it adds things to my To-Do lists. I want to have to be able to one-click an article and that’s why I built this feature. Other said they said things that alluded to the fact that they didn’t understand how the product worked. So they would say like, “Well the suggestions are all around the same topic and that doesn’t help me” whereas that’s actually a really good thing. And so I realized I needed to educate. And so there were three or four that were like that where I needed more education. I needed the people who were using it to understand how it works, to understand the benefit they could get out of it.

[34:55] So I listed these things out and one by one, I’ve been scratching them off and with the release of the one-click article feature, it’s done. The education I did through screencast during the trial and so people they had a few e-mails during the trial and I basically just had a question in there and would have – and once they answered and then the link went off to a screencast. Now since I started operation retention, my trial to paid conversion rate is up 100%. It has doubled.

[35:21] Mike: Wow.

[35:22] Rob: Yeah, in three months and I’m super stoked about that as you can imagine. And then the other thing the churn rate is one half of what it was when I started. So I’ve 4x’ed the funnel in essence.

[35:32] Mike: Nice, very nice.

[35:33] Rob: Yeah, I’m going to go in to more detail about it. I’m probably going to write maybe a book chapter under it. They just, you know, there’s a lot of stuff that I tried that works, some of them didn’t work. There are still tweaks to be made and it is kind of a minimum viable effort, an MVE at this point. Now, I’m just getting ridiculous with that.

[35:50] Mike: [Laughter]

[35:50] Rob: But no, it’s like some of the screencast, I mean they’re not like fantastic. They’re not well produced. It was me sitting there recording for forty five seconds and then just pushing it on to the server because I wanted to get the stuff done as quickly as possible. But even with that kind of home brood effort of really not having stuff professionally done and not spending a ton of time thinking it through, they had dramatic impact on, on these numbers. So, I’m super stoked. I’m also — I’m just stoked that it — I’m stoked that it worked. I’m stoked that kind of the hypothesis was validated. And just that, you know, all the stuff that we talked about it holds true, man.

[36:20] Mike: Very cool.

[36:21] Rob: You’re building the funnel, plug in the holes.

[36:22] [music]

[36:25] Mike: So if you have a question or comment, you call it in to our voicemail number at 1-888-801-9690 or you can e-mail us at questions@startupsfortherestofus.com. Our theme music is an excerpt from “We’re Outta Control” by MoOt, used under Creative Commons. You can subscribe to the podcast in iTunes by searching for Startups or via RSS at StartupsfortheRestofUs.com where you’ll also find a full transcript of each episode. Thanks for listening. We’ll see you next time.

Twitter Digg Delicious Stumbleupon Technorati Facebook Email

5 Responses to “Episode 83 | HitTail & AuditShark: Integration Marketing, Logo Design, Survey Analysis and Leaky Funnels”

  1. I love the podcast and the following is constructive criticism.

    Mike, I believe that basing all your effort on a survey is not the best way to go. People will say anything on a survey. People sign up to tons of mailing lists and doesn’t mean they will hand over their credit card. This is not data to base a business on.

    Rob, I believe you know this isn’t the best route and should hold Mike accountable. I’ve been listening to the podcast since like #25 and it’s been more or less the same stuff with AuditShark.

    In fact I feel Auditshark is more of a negative example, the exact thing to avoid when making a product: 1. making something huge that takes years to finish 2. starting without a real market (the “I think x or y people will use it” is not a real market), etc.

    What happened to shipping a minimum viable product? To being familiar with your selected niche? To having customers before even shipping?

    This is getting boring guys.

  2. Mike: Great job using 99 Designs – had a good experience with them as well. Would be cool if you share the top 10 or 20 designs with us. Keep up the great work guys!

    Jack C.

  3. Rob and Mike, You were talking about minimum viable things. Oren Eini also uses this term and is the owner of several successful software products. Oren is also big in the .Net community.

  4. I got curious and did a little searching for this minimum viable feature idea and found this (from 2008): http://www.netobjectives.com/blogs/more-insights-epics-vs-minimum-marketable-features


  1. Integrations: Simple Marketing for Developers | Zapier Blog | Zapier - Sync the Web - July 29, 2012

    […] integration with Highrise as one of the top things he did to increase revenue for BidSketch and in this podcast from minutes 28 to 32 Rob Walling talks about how well integration marketing has worked with his […]