Litmus: How A Secondhand Computer Turned Into A Multimillion Dollar Business

Paul Farnell is proud that he launched Litmus with a used computer and a few hundred bucks, instead of raising money for his company from investors.

But starting a business with little money meant he needed a lot of hustle, and you’ll hear him talk very openly about the feats he pulled off. I was going to add a few examples to this intro to give you a sense of how he did it, but I think you’ll miss the value if I take them out of context. Instead, I’ll tell you that if you trust me at all to steer you towards ideas you need to hear, you should grab this interview ASAP.

Paul Farnell

Paul Farnell

Litmus

Paul Farnell is the co-founder of Litmus. Litmus tests and tracks your email campaigns, so you can always put your best design forward.

 

roll-angle

Full Interview Transcript

Andrew Warner: Three messages before we get started.

First, do you know someone who wants to build an online store? Would you tell them that I said Shopify is the easiest way to build a robust store? When electric car pioneers, Tesla Motors, wanted to build their online store, they chose Shopify and so should you. What will you sell online after you discover Shopify?

Next, are you still coordinating projects by e-mail and wondering why you’re not productive? Check out TeamworkPM.net. TeamworkPM.net is trusted by big organizations like Universal Studios, right down to small two-man operations. How will you team benefit from a productivity boost by using TeamworkPM.net.

Finally, do you want to build a web app that interacts with phones? Twilio makes it easy. The founders of PagerDuty created a web app that tells site administrators when their systems are down. By using Twilio, they could send alerts via phone and SMS to their users. They didn’t have to build it themselves.
They could just use Twilio’s system. How will you use Twilio to build voice and SMS apps? Check it out at Twilio.com.

And here’s the program.

Hey everyone, my name is Andrew Warner. I’m the founder of Mixergy.com, home of the ambitious upstart.

How do you a side project into an over a million dollar a year business? Joining me is Paul Farnell, the founder of Litmus, a tool that lets online marketers test and track their e-mail campaigns. His business started in 2005 with just an $800 and, as we’ll hear in this interview, he turned it into a profitable bootstrap business. Litmus helps e-mail marketers see what their messages look like on 33 e-mail clients and mobile devices. Litmus also helps them avoid spam filters and enables them to measure their e-mails’ effectiveness. Paul, it’s good to have you on.

Paul Farnell: Hi, Andrew. Great to be here.

Andrew: All right, before we get into how you got here, how you built your business, what advice you have for other entrepreneurs, I want to get a better sense of what Litmus is and pass that on to my audience. So how ’bout an example of how a client of yours is using Litmus to improve their business?

Paul: Sure, yeah. Great question. We have a client, a fairly large company, who do a number of different services around e-mail for their clients. And they’re using our latest product, the e-mail analytics tool, to build a services package that they can sell to their clients. They’re charging $5,000 – $10,000 for this package, which previously didn’t exist. And that’s taken them, using our tool to get at the data, like half an hour, potentially, to put together. They collect all the stuff and we build the report and they present that to the client and so it’s kind of off the back of that client relationship that they already have.

Andrew: What kind of reports do you give clients? Open rates we already get through our e-mail systems and click rates, I know I use Aweber, I know a lot of people use MailChimp or any of the other services and they know what the click rates are. What other analytics can you possibly give us?

Paul: Sure, yeah. Good question. The way we describe it is we’re breaking down that open rate.
So of the people that have opened your e-mail, what do they actually do after they’ve opened it? You know with MailChimp or any other product, 40% of people opened your message. What we can tell you is of those 40%, what percentage of those kind of skim-read the e-mail or just deleted it outright? What percentage spent a few seconds reading it and how many people read it thoroughly? We can also tell you how many people printed it out after they opened it, if they forwarded it on to other people when they were looking at it. And so it’s all around that kind of level of behavior and engagement and tracking those details about the subscribers’ behavior that you don’t get with standard e-mail [??] stuff.

Andrew: OK. And what you are is another layer, another tool in addition to the software and the service that I use to send out my e-mail? You just enhance it?

Paul: Yeah. Exactly. You’d still want to use MailChimp or Aweber or whoever you’re comfortable with and we’re kind of just [??] with all of those and it’s just an extra level of detail on top.

Andrew: Who would have even known that there was even a business in this? I understand sending out e-mail is a business. I understand running websites is a business. I would never have imagined that analytics on top of an e-mail service would be a stand-alone business that would do this well.

Paul: That’s a good point. Obviously it’s not the only thing we do but we launched the analytics side about six months ago and that’s been really fast-growing for us. I think the simplest answer to that is just that people who are involved with marketing are obsessed with stats and anything they can get more figures on and learn more about their audience using, that’s something that they’re willing to pay for because you’re able to tie that down then precisely to the ROI you’re getting from their campaigns. If you’re able to use our product to segment your campaigns better and send additional messages to the most engaged subscribers and they’re buying more, then obviously that’s kind of a no-brainer in terms of a product that’s beneficial.

Andrew: That’s something that I’m seeing here in these interviews. For any product that’s out there, if you create an analytics package, that’s gonna be a need for it. You may not be the best in the business. You may not end up being profitable, but you are going to find a need. And so if there’s a new mobile device out there, if you create an analytics package for the mobile app makers you’re going to do well. If there’s e-mail out there, you’re going to do well by creating analytics for that. I’ll talk to you in a moment, actually later in the interview, about Chris from Wistia who has created a layer of analytics on top of video. Anything that someone created, if you add analytics on there, you’re going to address a need.

All right. So let’s go back and understand how you discovered this business and understand, more importantly, how you grew it. So going back to 2005, what was the original opportunity that you indentified?

Paul: Yeah, that’s a great question. It was myself and two other founders, and we were doing freelance web design work for clients. It was actually while we were still in college. As part of building these sites and blogs and so on for clients, we needed to do cross-browser testing, and Litmus actually started out — it was called Site Visitor at the time — as a cross browsing tool. At the time, there were a couple other around. There was BrowserCam and one or two more, but they were actually relatively expensive at the time and the service that they provided, and the user interface and the speed of the tests and this kind of thing were pretty bad.

So we just kind of thought, “Well, this is something we have to do day to day and looking at the products out there, I think we can do a better job.” Actually, the first prototype of Site Visitor was built in a weekend and even when we launched, it was running on two or three old computers that I’d got from my dad’s company running under a desk at college on the university network. It started out really small. It was really just answer that need that we had and that feeling that we thought this could be done much, much better.

Andrew: How did you even see that there was an opportunity in designers needing to see what their websites looked like on different browsers? How did you come up with that opportunity? How did you identify that problem?

Paul: That’s a good question. I guess it was borne out of our own frustration. Cross-browser testing has always been a really pain to do. There’s more tools around there to make that easier now, but in 2005, there were fewer. And so it was just through our own experience that we thought that we would definitely pay for something that did this, that made this easier and that was nice to use and was actually fast and was going to save us a lot of time.

Andrew: What were you doing?

Paul: What were we doing?

Andrew: Yes.

Paul: We tried BrowserCam, but that was kind of a little bit too expensive.

Andrew: I’m sorry, no, what I mean is, I know it was three of you co-founders who launched the business and you identified a problem because you felt it yourselves. What I’m trying to understand is, what’s the business that you were in that helped you experience this problem? Were you designing webistes?

Paul: We were, yeah. We were building sites for clients. Yeah.

Andrew: And so you wanted to see what your sites looked like on Internet Explorer, on Firefox, on Safari. And how did you do it? You used the competition?

Paul: Yeah. We tried the competition and found that very frustrating and actually ended up just doing it manually instead. But that’s much more time consuming than having a service that would do that for you.

Andrew: And manually means going over to a Mac and seeing what your websites look like on a Mac and firing up Firefox and so on.

Paul: Exactly. Yeah, and then if you don’t have one of those platforms, it’s calling up a friend and saying, “Can you load this up?” And then you’re on the phone with them or something trying to get them to describe, well is the footer lined up with… and so on. It was just a nightmare and obviously not the best way to do it.

Andrew: I’ve been there. I’ve done that via e-mail with a few of my users on Mixergy.com trying to identify why they couldn’t see my site properly on their computers, on their browsers, on their operating systems.

OK, so when you say that you built the first version in a weekend, that tells me that you didn’t build a very broad product. You built something that narrowly focused on simple issues or on a few core issues. What were those core issues that you focused on?

Paul: That first version simply let me enter in a web address and see a screenshot of the site on two different browsers, I think it was initially. And that was all it was.

Andrew: Just two?

Paul: Yeah, for that first version. I mean, it worked. It was a little bit slow. It wouldn’t work on every site and, you know, if you had Flash or something, it would crash. There were a multitude of problems with it, but that whole thing, what we wanted to do was a proof of concept, really, to see that this was possible.

Andrew: OK. Was there any plan to sell that first version that you launched over a weekend?

Paul: I’d say no. That was definitely intended as a prototype, although, to be honest, it was just expanding upon that is exactly what we launched, expanding that to eight browsers, scaling it up from just testing one site at a time to testing two or three simultaneously initially, and that was it. The code that was written for that prototype was absolutely launched into production. I think the initial idea was that we would kind of rewrite it, but in the end there was no need to and we just built on top of it.

Andrew: How do you build something like that? I mean, the first simple version, how does it know what a website looks like on two different browsers?

Paul: It was me that kind of cobbled this together and we used all these Windows Marcos is how it initially worked. We would script the browsers to open up this web address that it got from here and then take a screenshot and save that here and FTP it here. It was very, very, very basic and a little brittle initially.

Andrew: I see. So you weren’t duplicating the experience of watching a website on Internet Explorer, you actually fired up Internet Explorer on a Windows computer, took a snapshot of it, saved it onto your hard drive, then moved that to an FTP folder and then displayed that file on a computer screen. That’s what the process was?

Paul: Exactly. Yes.

Andrew: Is that what still happens to this day?

Paul: Gosh. The honest answer to that is that I actually don’t know. Now we have a development team who are way more skilled at this than me. I mean, I understand broadly how everything works, but exactly how the browser testing works now… We’re still loading up actual browsers and taking screenshots of those and delivering them, but there’s a whole queuing system and there’s S3 and there’s all kinds of other things to make it work much more quickly than it used to.

Andrew: I see. OK. Earlier you said that we used Macros. To me, that indicates that you’re not a developer, right?

Paul: Yes. So me personally, no. I mean, I can kind of hack together little bits and pieces here or there, although I’m not allowed to do that anymore. I would build sites for clients and maybe I needed to do a little contact form script and that kind of thing. Personally, I’m not a developer and that’s where the other two founders kind of came into the picture, because they are and were able to take what I built and scale it up to being not just a prototype, but something that could actually be launched.

Andrew: What kind of software do you use to create a Macro that does what you described earlier for the first version?

Paul: That first version? I don’t know if it even exists anymore, but we used a product called Macro Express, which worked fine, I guess. You wouldn’t want to run the volume of tasks we do now on that kind of system, but it was fine for the first.

Andrew: Sorry to stick on this for so long, but I’m just curious because I’ve never seen anyone build a first program using a Macro program. I don’t remember Macro Express well, but I think the way it works is that you essentially do an act, it records it, you can then edit the steps in that process, but that’s what it is. The initial creation of the Macro isn’t you typing in code, it’s you doing the action on your screen, it records it and repeats it and you can edit it. Did I get that right?

Paul: That’s absolutely right, yeah. And I suppose just to touch upon something else, what perhaps sets Litmus apart a little from a lot of other web apps is that we don’t just have the web application itself, which nowadays is Ruby on Rails, saving and loading things from a database and responding to requests and that kind of thing, but this whole backend part of the system, which is doing all this automation with e-mail clients and all that kind of stuff, which is actually multiple times larger, in terms of the code base, than the front end application itself. So to tie that back to our initial prototype, the Macro Express software was doing that backend processing of loading up the browser and taking a screenshot and then there was a small amount of VBScript, like ASP code, that would just submit the form and start the Macro going and save it into the queue and that kind of thing.

Andrew: OK. I’m not a developer myself, but what you’re describing is something I could even put together.

Paul: Oh yeah. Absolutely. I mean, I’m not a developer. It was very straightforward, that initial version.

Andrew: So you put this out into the world. It only has two browsers. Who do you show it to and what’s their response?

Paul: We did add a few more browsers before we showed it to the whole world and we also stuck the payment processing stuff in there, but it was never an initial freebie or anything like that. It was charged for from the start. I went to a conference in Copenhagen called Reboot.

Andrew: Actually, I apologize, Paul. I’d like to stay and get a lot of detail in the early days and then move on. I mean that first only two browser prototype that you built over a weekend. Did you just show it to the three of you and say, “Wow, this thing works. It’s kind of cool” or did you show it to some friends, did you show it to some clients? What did you do with that initial launch?

Paul: Yeah, that’s a good question.

Andrew: I shouldn’t even say the initial launch, I should say the initial creation, the initial thingy. What do you do with that thingy?

Paul: As far as I remember, I obviously showed it to Matthew and David, the other founders, and close friends and such. But also, I was on a major mailing list of web designers in the UK and there was probably 40 or 50 people on there and I’d been on there for a number of years. I’m pretty sure that I sent the link to there first and that’s kind of like your initial test of the market. These were all people who were perfectly targeted to buy this if we were going to launch it, albeit perhaps in a slightly fuller-featured form. And so it was good to get that initial feedback from them and close friends before we took the step of adding the payments and scaling up a little in terms of actually launching the product.

Andrew: I see. You know, I’ve seen people do that on Hacker News and on message boards. They say, “Hey, I built this over the weekend. What do you think of it?”

Paul: I think that’s a really good way to… In this context, it was more of a closed community. It wasn’t up on Hacker News for tens of thousands of people to see, but just to get very focused feedback. Very focused in terms of the type of people that we were asking about it. They were 100% the perfect audience for this.

Andrew: So tell me about the feedback that you got on this initial creation. What was it?

Paul: Gosh. If I’m honest, I don’t remember it in detail, but I’m sure it must have been positive in order to encourage us to build the next version. But we weren’t riding everything on what would people on this mailing list say. We were quite headstrong and we thought it would be fun, at least, to launch this as a product and have that experience of selling a productized thing as opposed to what we were doing already, which was services, essentially.

Andrew: I see. So this wasn’t a post to find out whether you should go or not with the product. You were going to go one way or the other. You were just kind of showing off what you created. Is it fair to say that you were also looking for a little bit of encouragement?

Paul: Definitely, yeah.

Andrew: I see. OK. When it’s just in your head or when it’s just on your own computer screen it’s not as exciting as when you see people interact with it.

Paul: Yes. Absolutely. Yeah.

Andrew: OK. Based on the feedback that you got and the experience you had using that first product, what did you decide that you needed to add or change before launching?

Paul: Yeah, good question. We had a huge list of stuff that we thought it would be amazing if this tool could do. And actually, we ended up, on our website when we launched, listing all the stuff that was coming soon. But in terms of things that it absolutely had to have, we wanted to keep it at an absolute minimum. And so, we added a few more browsers, because obviously that’s kind of the whole purpose of the tool is that it tested some range of browsers. And aside from that, just payment processing, basically. Just payment processing and a few more browsers. We refined the UI a little bit. It didn’t even have a login system initially, so we had the users accounts.

User accounts, payment and a few more browsers. That was it, really. And then everything else, we just listed as things that we were working on and were coming soon just to kind of get people excited and get some feedback from the people that saw it after we launched. Which of these five different things that we think are cool are you most excited to see? And that’s what we’re going to work on next.

Andrew: Why was money, why was charging so important to you?

Paul: Good question. I had been building stuff and selling stuff since I was 11 or something. It’s just kind of an obsession of mine. I used to write really simple pieces of software in Visual Basic and Delphi and sell them on the shareware sites. To be honest, personally speaking, I see purchase prices as validating that something is worthwhile. If something is completely free, sure you might get 1,000 people using it, but do they care that much? Do they care enough to actually part with money? I feel like that adds value and adds proof, to me personally, that this is something that’s worth doing and that people actually really care about and that it’s benefitting them enough to an extent that they’re actually handing over their card details to continue using it.

Andrew: I’ve seen that quite a bit in my interviews. I’m wondering why. Why is it that having fewer people who are more into your product, in fact, so much into your product that they’re willing to take out a credit card. Why is that better than having more people who are using your product on a more frequent basis because it’s free? Why is one better than the other in your mind? Just for you.

Paul: I guess I am much more… I suppose it comes down to that sense of sustainability and, to be honest, security in a sense. So if you’re not charging, it’s costing you something to run it and that money is essentially going to be coming from an outside investor and that’s something that I’m personally not as comfortable with as something that I feel that we have complete control over and can grow in a more sustainable way. And then you feel you have more of a relationship with those people that are using it. There may be 1,000 people using this free site, but they could just go somewhere else within a day. They don’t care about it the same extent or have the same relationship with the founders or the product creators that you do if you’re invested enough to be paying for it, in my opinion.

Andrew: OK. I understand that. What I don’t understand is this. In the 37signals article that I read about you, you brought up the fact that you used old computers. In this interview you brought up the fact that you used old computers. Why is that so important that you bring it up so much? What is it about these old computers that’s so significant for you?

Paul: I think it kind of epitomizes how we saw things when we started, to be honest, how we still see things, in terms of being as frugal as we can and being able to build something from… These old computers had almost no value, yet we were able to use those to launch a service that had 100-odd people that were paying for it within a few months and I guess to the outside world looked just as important as a big company’s product. And in fact, we were, I felt, doing a much better job of delivering that service using this old hardware and our spare time whilst we were at college than an established company that had been around for a number of years, building a similar thing and selling it to the same audience.

Andrew: I see. OK. All right, I understand that. I understand that pride. I take that kind of pride in running. I love that I have older sneakers, that I have not the right outfits and that I run more often and run faster and further than people who have these perfect outfits and all they seem to care about is having the right equipment.

Paul: Yes.

Andrew: It almost challenges yourself.

Paul: It does, yeah. I remember… Well, anyway, yeah.

Andrew: No, no. Share it. I’m here to hear your story.

Paul: Not technology related, but exactly as you said, I used to do a lot of cycling and I once entered this cycle race not really knowing exactly what the deal was. Anyway, I turned up at the circuit and everyone else in the race had these amazing racing bikes and all the lycra gear and this kind of thing and I was there with like this muddy mountain bike, wearing just a t-shirt, just jeans and stuff. That’s similar to the running.

Andrew: You know, I’ve gotta stay on this for a little bit longer. One of the reasons why I started doing these interviews is to understand what entrepreneurs and successful people are really like, not what the self-improvement books say they’re like. The self-improvement books often will say that entrepreneurs and successful people in general have, what do they call it, not a mantra, but something that they say over and over. Affirmation. They say affirmations. I haven’t talked to an entrepreneur here who does affirmations on a regular basis, but I kind of see a version of it in the things that we do that tell ourselves we are great and we are meant for greatness and we are going to achieve greatness.

Things like using a computer that’s crappier than the eight year old down the street has and by doing that, you’re saying to yourself “I am such a good entrepreneur that I can build an incredible company even with this.”

And, in your case with cycling, not having the right equipment or the perfect equipment, you’re saying to yourself every step of that journey, “I can do this so much better.”

Do you agree with that? Or am I just trying to connect two things that shouldn’t be connected? What do you think of that?

Paul: Interesting point. I think that’s definitely true and I think there’s definitely that sense of pride, almost in challenging yourself. To be honest, we all know that the tools that you have to build something are not what’s going to make that great. And I think that’s important to remember. And I think there’s a certain amount of focus. You know, focus on the right things.

You’re not focused on “Oh, well, I can’t really start building this because I haven’t got a Mac” or something.

Or, “I’ve just got this old machine. I can’t afford this great server that I want to run it on because it might have 100,000 users next month.”

It’s focusing on just trying to build the best thing that you can with what you have and having confidence in yourself to be able to achieve that without needing, well, to be honest, outside funding or great, amazing, brand new, sparkling tools and all that kind of stuff. I definitely agree with that.

Andrew: All right. So you improved on the first version by adding more browsers, you improved by charging and you also had a “coming soon.” What was on your “coming soon” and why did you include this publicly?

Paul: That’s a good question. On the “coming soon”, it’s funny because a lot of the things that were on the coming soon are still coming soon or we never ended up building. We wanted to do colorblindness testing, so we would show you your site as it appeared to people with the three most popular types of colorblindness. We had a screen reader test, so we would actually deliver an MP3 of how your page was read by a screen reader. What was the other thing? This was cool, but I don’t think it applies these days so much, actual videos of your site loading at different connection speeds. So you were on broadband but you can watch a video of how your site loads, in what order do the images come in and this kind of thing on a 56K modem. I’m trying to think. And then I guess we mentioned having a wider range of browsers. There were a few extra browsers that were on that list as well. But yeah, it was the colorblindness, the loading videos and the screen reader and actually, we never built any of those.

Andrew: Why not?

Paul: I guess because, by the time we got it out… These were the things that we thought were cool and we thought people were going to be excited to use and, to an extent, people thought they were cool ideas, but they would say, “Yeah, they would be good, but what we really need is this other thing,” which we maybe had considered but had ruled out as important.

Andrew: Like what? What did they say that they wanted that you just didn’t include on your “coming soon” list?

Paul: Perfect example is e-mail testing. Essentially now that’s kind of 80%, 70% of our entire business is from e-mail testing, but that wasn’t on our “coming soon,” that was something that we thought, “Oh everyone’s going to love it when we do this.”

But it was something so many people said to us, “Oh, well, cross-browser testing is hard, but what’s even more difficult is e-mail testing.” And once enough people said that, we were like, OK, maybe we should not worry about the colorblindness thing that a few people thought was neat and we should do this thing that dozens of people have said would be super useful.

Andrew: What are the benefits of posting online and telling the world what’s coming soon?

Paul: I guess, initially, we were doing that to try and build excitement about the product and probably also, if I’m really honest, because we… I wouldn’t say ashamed, but we were a little embarrassed by how much of a minimal final product this really was. Nowadays, we don’t generally talk about what’s coming soon. If a customer e-mails to ask, “Oh, are you going to have this mobile device for testing?” We’ll say, “Oh yeah, we’re working on that. It should be there in a month.” But we won’t post on our site or on our blog or maintain a “coming soon” list now of what we’re working it. We just kind of launch things as they’re ready. I think it came out of that, honestly, a feeling that maybe it was a little bit too small, but if you kind of go with us now, we already are beginning to work on these other things. We obviously ended up scrapping a lot and built something slightly different.

Andrew: There was an advantage that came with it, which is that you got feedback from people. Did you actively hear from people that they didn’t want things like the MP3 reading of the text on their website and the colorblindness? Did you hear back from them about that?

Paul: You know, in honesty, I don’t think we ever got actual negative feedback about those things that were on the coming soon list. And I guess, really, that’s largely because the people we were talking to were technical kind of web people and, even now, I think that sounds like a really cool idea and if somebody said that to me, I’d say “Oh yeah, that’d be really neat. I’d like to see that.”

It doesn’t necessarily mean that they should go ahead and build that. So, I don’t think anybody ever said, “No, that’s a poor idea. You shouldn’t have that feature.” It was more that we got way more positive ideas for other things that we should build that far outweighed the number of people that said, “Oh my God, this MP3 thing would be amazing.”

Andrew: I see. All right. What about sales in the beginning? How did they come in and how many sales did you get?

Paul: I went to this conference in Copenhagen and this is where we kind of launched. We did a promo deal for, I believe it was, it was $19 or $29. It was $19 and the actual price was $29 a month, but we did this kind of pre-launch — we called it a pre-launch, but I mean, we’d launched – deal for $19 a month. And basically, we printed off 200 of these cards that had a code on them and an address to go to and I just went and pitched this to every single person I met at the conference. And as a result of that, I think within a couple of weeks of getting back from that conference, we had the first 20-25 people who had actually signed up and begun paying for it, which was incredibly exciting because this was, essentially, two weeks after launch.
We launched for this conference. It was the first time that we’d really gone out and pushed it. That was incredibly validating to have people who were willing to being paying right away. And a lot of that was excitement around stuff that was perhaps coming soon, the utility of the product initially, obviously and also that we had this discounted price that was only going to be valid for… You needed to sign up in that first month in order to lock in at this $19 a month mark.

Andrew: And you didn’t have a booth at this conference. You didn’t have an official role. You weren’t a speaker. You were just a dude walking around with your coupons handing them out to people one at a time, describing what your product does and hoping that they’d take the coupon, go online, type in the coupon code and get your product at $20 instead of $30.

Paul: Yeah, exactly.

Andrew: And that was a one-time fee or a monthly fee?

Paul: That’s a monthly fee.

Andrew: It’s a monthly fee. Oh, wow. Why did you make it into… I’ve gotta come back to the 25 people who signed up in this, but first I need to understand, why’d you go monthly instead of one-time or per instance?

Paul: Yeah, because we thought that was a nicer revenue model for us to have and because the use case we’d envisaged and that we personally fitted into was that of a freelance designer, or a few people working together, on various web projects and they’re doing this as a career, maybe it’s part time or it’s full time, but they’re doing regular projects. They’re doing one or two a month or something like that, building a site or two a month and, presumably, they need to test every single one of those. That’s who we’re aiming at, these people that do this for a living, as opposed to, I don’t know, like a business owner who’s heard a report from a customer that they couldn’t check out in IE6 and they want to go check out what the checkout page looks like in IE6. That’s not the people were targeting, and so it felt to us like that monthly model fit in quite well with that and it also fit in quite well with what we wanted to have as a business.

Andrew: I could understand, actually, why targeting designers, people who keep working with web pages would be a better target audience. I’m wondering why you picked them. Did you pick them because they’d be better, because they’d be easier to find ,because they’d be more likely to buy, because they’d be able to pass the expense on to their customers? Did you pick it for business reasons like that or because that’s who you were exposed to and that’s who you knew?

Paul: I think it’s both. It’s mainly that’s who we knew and, from a business point of view, that’s who we could speak to, speak in their language and that’s who we could reach. I knew all of these web design communities and design sites and had been active on there and on these different mailing lists and forums and stuff for years, and I saw that, from a marketing point of view, that was going to be a group of people that we could easily reach out to versus if it were for business owners with problems with their checkout, that would have been a more difficult place for us to start, in terms of how we’d be able to reach them.

Andrew: 25 people. That’s not a lot to start off with. What did you learn from trying to recruit people? What’d you learn about your product? My sense is what you learned is probably more valuable than the revenue you got from 25 people.

Paul: Yes. It was out of those initial 25, and other people that had tried it briefly for free, or whatever, truly pre-launch, was the suggestion to do e-mail testing.

Andrew: So from talking to people one-on-one, that time, that’s when they said, “No, give me e-mail testing.”?

Paul: Yeah. I believe a lot of them had used the product a little and it would be, say a week or two after that, we would speak to them again and either I knew some of them already, and some of other people I knew in Manchester, where I was based, had signed up during that pre-launch period as well. But yeah, it was meeting those people in person and talking to some of the other people who had signed up. That’s where that initial idea of e-mail testing came from. And so, in terms of value, that’s millions of dollars worth of value from that suggestion. So certainly that was more valuable than the revenue, but I wouldn’t underestimate. I know 25 people doesn’t sound like a lot now, but at the time we were blown away. We were so excited to have those 20-odd people on board. It was crazy that people were actually handing over money and committing to a monthly fee to use this thing that we’d only finished a week or two before. It was kind of mind-blowing at that point.

Andrew: Ha ha. All right, actually, I shouldn’t minimize that. That’s $500 a month that’s coming in from paying customers for a product that you just recently launched. That’s very impressive.

Paul: And that’s essentially covering our expenses. I mean, we weren’t earning very much, but just in terms of the hardware costs and that kind of thing. They were almost nothing because we were running it on the university network and our hosting of the site was like $20 a month. It was good.

Andrew: Now before I continue with your story, I gotta make this observation. You’re saying you’re in the design world, so I start paying attention to the design of the room around you, to the way you’re dressed. You’ve got a good haircut, great haircut and a tie. What kind of Internet entrepreneur wears a tie? Is it because you’re doing an interview here on Mixergy?

Paul: No.

Andrew: You wear a tie regularly?

Paul: Yeah. I mean, not every day, but yeah, once or twice a week or something. Yeah.

Andrew: Why? Is it just for your own design or do you need it for customers?

Paul: Oh no. Just for me, personally. It’s not a customer-facing…

Andrew: OK. So you’re a person with a good design eye. That’s what I’ve read about you online. That’s your passion.

Paul: That’s true.

Andrew: How does someone with that kind of passion get into business? Why not focus on design? I guess you’re saying that you always had an interest in business, right?

Paul: Definitely, yeah. Absolutely. From making little sandpaper things and little point of sale stuff to sell in a shop across the street from my house when I was 10 or something, from that point forward, I definitely had that kind of interest in business and that sense of making money and building something to sell. There’s a family emphasis on that. My father ran his own company the whole time I was growing up and that was definitely an influence on that as well. I care a lot about design and the user interface of the product and design in general. For me this is kind of a perfect situation. I still do the UI design for Litmus and design our site and everything, so I get to do that, but also have the satisfaction of running something that is profitable and something that, again, going back to that sense of validation, of having people hand over money and get that feedback from people that this is actually changing the way they work, which is incredibly satisfying.

Andrew: All right, continuing with the story now. You get all this feedback telling you to focus on e-mail. How soon do you start adding e-mail.

Paul: Let’s see. It did take us a little while because, as it turns out, testing e-mail clients, although it seems similar, you’re taking screenshots of something and delivering them to the customer, is like ten times harder than testing browsers. In the end, what we actually did is we got the… We’d been working a bit here and there, and this was actually mainly Matthew that was building this, and we got the three of us together. I had moved to London by this point and we’d graduated college because we started right at the end of college. I’d moved to London and the three of us got together for four or five days and we were just like, “Right, at the end of this, we’re going to launch e-mail testing.” And it was just a lot of Red Bull and very little sleep and just this crazily intense period of time. And at the end of that, we did have something that pretty much worked and so we launched that in beta. Realistically, I’m just trying to think back on exactly how long this would be. You’re probably looking at 9 to 12 months since we had that initial pre-launch phase until we added the e-mail side of things. So it did take a while.

Andrew: Why?

Paul: We were still, at this point, working on client projects at the same time as Litmus and so we weren’t able to focus all of our time on it, which became gradually more and more frustrating and it became clear that we weren’t going to really be able to grow this is a proper business unless we were able to give it more focus. I guess that’s kind of a tangential story there and how we solved that. But yeah, we weren’t able to dedicate all of our time. Matthew and I worked on a pretty big project for a client that ended up taking a lot longer that we thought, and kind of stressful and we were graduating and that kind of thing. So that’s why it just kind of overran, really.

Andrew: I have too big of an agenda with you to spend too much time on that transition, but I’m curious. You said that there was a story about that. I’d love to hear just a little bit of it.

Paul: Yeah, sure. We were getting more and more frustrated about not being able to focus on Litmus full time and it actually took us a while because although we were frustrated, we were still nervous. We’re thinking, “Well, if we stop doing client work, then we have no income. The income from Litmus right now is not enough that we could pay our rent. We can’t divide that be three and have enough money to live, so what do we do?” In the end, I took a loan just to kind of fund us for three or four months or so to give us time to really work on the product.

We actually rebuilt the whole application, the frontend part of it during that period, redid the site, redid the design and had a big marketing push with this idea of patron accounts. We knew that we needed more money than we had in the bank to get us to the point where we were profitable in terms of paying our own salaries, or paying us enough that we could just pay our rent and eat. We were able to use that three months or so that the loan gave us to build the product and do a lot of marketing, but we didn’t have quite enough and so we came up with this idea of patron accounts, which is actually what ended up tiding us over and making us profitable.

We said to all these people, especially reaching back to those first 25, and all the people that had joined us within that first year or so, “Look, we really appreciate your patronage so far. We want to build all this other stuff, but we don’t have the money to do it. We want to be able to do it quickly and you’re going to love it and here’s what we’re going to build. And so, commit to an account for two years.”

I can’t remember now what the price was, but it was like $300 or around that kind of mark.

“So you can get this amazing deal. Two years worth of service. You’re going to get this badge you can put on your site to show you’re a Litmus patron.”

We had a limit of 75 of these accounts.

“We’re going to give you complete features before everyone else.”

I mean, we just had this whole list of stuff.

“We’re going to give you everything we can.”

And that sold really well. So that gave us this kind of lump sum income within like a week of money that, after the loan started to run out, tided us over until we were profitable and had enough revenue coming in from the monthly customers.

Andrew: What did it feel like to say, “We don’t have enough money” to say it publicly?

Paul: We worded it quite carefully. I should have gone back and found that blog post of exactly what we said. It wasn’t that if we didn’t get this money everything was going to shut down. It would just be that we wouldn’t be able to work on it full time, so the development would have been slower. So it wasn’t one of these situations where you see other startups that are almost out of money and asking for donations and this kind of thing. It wasn’t like that. I forget the exact wording, but it was more positive about we want to be able to launch all these new feature and we want to be able to do it faster and we need a little investment in order to be able to do that and we want to take that from you, the people that supported us so fast and we want to give you a great deal and reward you for having been a part of this for the past year and a half, or whatever it was at that point.

Andrew: Do you remember how much money you got from that?

Paul: Off the top of my head, I don’t I’m afraid, no. It was about the same, I think, as the loan. Because our initial loan was for £10,000, that’s what we took to kind of get us that few month runway. I think we ended up selling 50-odd of those accounts, so we didn’t actually hit the 75, but we sold quite a few. And a lot of the people, of those first 25 and of the patrons obviously are still customers of ours now, which is awesome.

Andrew: That is awesome. What were you selling the patron accounts for, roughly?

Paul: I believe it was about $300.

Andrew: $300. OK. Wow. Congratulations. So now you get to focus on your… I’m glad that I asked that. So now you get the money, you get to focus on the developing a product. You use that money and the time that you have to build out the e-mail portion of the business.

Paul: Exactly.

Andrew: What’s the customer reaction to it when you launch it?

Paul: The e-mail thing went really well. We got a lot of coverage from blogs and stuff like Campaign Monitor, who I remember just being so chuffed when they wrote this article about our e-mail testing and how all their customers should use this. Because they have their own e-mail testing thing now, which [??], but at the time they had nothing but they completely understood that as being a huge need and something that wasn’t really serviced elsewhere unless you were willing to pay thousands a month, basically. And so they were really, really on our side and that was our perfect audience because now we’re selling browser testing and e-mail testing. The customers of Campaign Monitor are very much web designers, people that are also building sites for clients, they’re freelancers or from smaller companies and so that was the perfect audience for us and we got a huge lot of sales right in that first week from the coverage we got from them and from a few other places. So it was definitely, I’d say actually a bit bigger reaction than we expected on the e-mail testing side, which was really, really exciting, really validating.

Andrew: What was your biggest source of revenue, not revenue… What has been your biggest source of customers?

Paul: I guess that’s clichéd to say it, but word of mouth. We don’t do a great deal of advertising online. We kind of try things out, but it’s generally people get referred by other customers or, a particularly happy customer or maybe a partner of ours, like Campaign Monitor will write up articles about us or about a new feature. Nowadays, we try to do a lot more of our content, of our own research that appeals to e-mail designers specifically. We just hired somebody whose focus is essentially that, just doing that research, writing up that content report and that kind of thing and having that shared around.

Andrew: I see. The way that you’re planning to market is just to release research, to blog it, to create reports and have other people come check, have the target audience of designers come check out the reports and information on your website and maybe start publishing articles about it.

Paul: Yes, exactly. Yeah. We’ve done that to a smaller extent over the past few years, but essentially that was my job and it’s hard to find the time to do that as frequently as we would like which is why we hired somebody to…

Andrew: So aside from word of mouth, is that the best source of new customers for you?

Paul: Yes, definitely. Yeah.

Andrew: All right. Let’s see what else I’ve got. So now we have the e-mail product, we have the web product. Is the web product called Alkaline?

Paul: Alkaline is a desktop version of it, essentially. But I guess it’s worth mentioning at this point that we have, you know to use that phrase, pivoted from the browser testing side of things very much to e-mail. So if you look at the site now, there’s very little mention of browser testing, although that’s still a feature. We kind of focus on that as landing page testing now, because we’re in the e-mail marketers. So there’s like a follow-on test from testing your e-mail kind of thing.

Andrew: Why isn’t there enough of a business in webpage testing? I would think that more people are doing webpages and marketing on the web than are marketing via e-mail.

Paul: Yes. I mean, I guess that must be true.

Andrew: Yeah, that is true then I would think that there’d be more of a need because of that. Why isn’t there more of a need?

Paul: I think there’s a few things that changed since we first launched and it mainly revolves around competition. So, you’ve got things like Adobe BrowserLab, which is a great tool and it’s completely free and it’s really quick and it looks nice and it does a good job. It doesn’t have all the features that Litmus has from a browser-testing point of view, but for most people, it’s probably good enough. There’s actually now there’s like a dozen different startups that launched in the last year or two who are doing webpage testing and are doing fairly innovative things. They’re trying to automatically detect problems and highlight particular browsers that they can tell have aligned something incorrectly or whatever it might be. So I definitely wouldn’t say that there’s not money in building services like that, but I think it’s getting harder and I think that that’s also not helped by the fact that the new browsers that are coming out, IE9’s great, Firefox has always been very good, so many more devices and browsers using WebKit, like Chrome and that kind of thing, that it’s becoming, to be honest, less of an issue. I think there’s

Who should we feature on Mixergy? Let us know who you think would make a great interviewee.

x