Techzing 45 – Taylor Norrish / PrintFriendly and GovIt

Justin and Jason interview Tayor Norrish about his startup experiences with PrintFriendly and GovIt. Specifically, they discuss how he came up with the idea for PrintFriendly and how he found a developer, the algorithm used by PrintFriendly to parse and tag relevant content in HTML pages, how GovIt got started and why (to his regret) he declined the first two and only acquisition offers, the importance of finding a partner with complementary skills, the relative cost of various web technologies, comparing the cost of Java and Ruby developers, the importance of experience vs passion when hiring a developer, the difficulty of finding good designers and developers, Elance and ODesk, targeting a niche market vs a horizontal market, the theory of the cargo cult, using Google Optimizer for A/B and nth degree testing, starting a service based on the Farmville concept of building something and watching it grow, brainstorming a pricing strategy for Swarm and Digg’s recently announced change of strategy.

15 Comments
  1. Bopinder Abu Morpalinder Singh says:

    Nice. $2k/month on ad sense. How much money did he invest in printfriendly again?

  2. Charlie Irish says:

    Great automated business model – get a few of those going and you’re laughing.

    RE: Swarm app
    Justin, people are always going to find a way to pirate your app (e.g. cydia or UDID Faker) so you need to roll with it. The 50 plays for free sounds like a great idea to me but you could also try to get a recurring business model: by getting people to do in app purchases to pay for another 100 games. This might increase the total potential revenue.

  3. Julian says:

    It’s funny, I was thinking about how you would ‘feature limit’ chess just the other day. Here are some thoughts that may relate to Swarm. In part, these are for player v’s computer as opposed to player v’s player:

    1. Limit the computer’s playing strength. Especially for Swarm, everyone is coming in as a novice.
    2. Lock the ability to pause games, take back moves, hints etc.
    3. Lock any analysis tools. e.g. Pychess has a nifty graph that shows who is in front and by how much at any given move.

    All of these relate to how serious the player is, so they can’t hack their way around the limitations and it doesn’t really cause pain until the player is serious enough to shell out for the paid version.

    By the way, ChessMaster by Ubisoft is a brilliant example of how to make a strategy game more human and fun. It’s clunky and a bit ugly, but things like having ‘personalities’ to play against instead of just a playing strength makes for a more engaging experience.

    Good luck with it, it sounds like a pretty cool game.

  4. Justin says:

    @Julian – Thanks for those good ideas. I think you’re right because Swarm is a completely new concept I will probably have to evolve how to price it and deal with the upgrade path. I guess the most important thing I need to know is… do people think it’s any good and worth playing! If yes, then it has a future.

    @Charlie – Yikes! UDID Faker. Well the hope is that not too many people use that kind of thing….

  5. @Bopinder Abu Morpalinder Singh
    About $16K. Mostly development. I’m not a programmer.

  6. soitgoes says:

    Another really interesting guest. Your podcast is my current favourite (used to be stackoverflow). It’s great how you encourage your guests to talk in detail about their startups including the actual financial figures. Very educational and inspiring stuff! Thanks guys.

  7. Robin says:

    Another good show, it would have been nice to here a little more from Taylor but enjoyable and informative non the less.

  8. Justin says:

    @Robin, you really made me laugh with that one! LOL 🙂

  9. Jason says:

    @soitgoes Yeah, Stack Overflow used to be my favorite podcast as well, but after a while it started to get a little repetitive and then it just died. ;( Oh well, I guess we’ll just have to pick up the slack. 😉 Thanks so much for the positive feedback!

  10. Taylor was a delightful guest. I liked how it wasn’t 100% an interview of the guest, but a mix of interview + the usual stuff, with the guest’s observations and opinions thrown in for good measure.

    I agree with some of the comments above that Techzing has become the podcast that I look out for and get happy when it shows up each week. Keep up the good work, guys.

  11. Jason says:

    @Aaron@Phrasemix Thanks so much for the positive feedback! It feels great to hear when people are enjoying the show.

  12. Wow as I am sitting to comment on this episode it looks like you just posted the new show. I guess I know what I’ll be listening on my commute tomorrow! 🙂

    I enjoyed the show with Taylor. He actually seemed to have helped balance the energy flow between the two of you!

    A comment on “don’t do things in ASP.NET”: I think it’s a fine platform for internal business web apps especially when you need to interface with a myriad of systems over SOAP, XML over HTTP, message queues, etc. It can also be ok if used judiciously in a sizeable company (e.g. at my work). But I agree that if you’re an entrepreneur, unless you already have some deep .NET devs it could be challenging to deploy and scale. Although Microsoft Azure seems like it will eventually be a solid platform for deploying and scaling .NET sites to the cloud, it is still in its infancy. So maybe in a year or two it will be ready.

    In the same vein, it is true that the cost of experienced .NET or .Java developers will be reasonably high especially if they are enterprise developers skilled in the full breadth of the platform. I suspect that the cost of similarly skilled Ruby developers would also be high if they have worked on scaling, distributed databases, sharding, queuing, CRON and so forth. At the same time for a startup, it would be easier to leverage more affordable and maybe less experienced Ruby devs.

    On the topic of finding good designers, I have literally just been through that and its tough especially because: visual appeal is very subjective, designer portfolios can’t capture the “feel” part of the look and feel. That is it is difficult to gauge what usability expertise a designer has. Also it is challenging to estimate the cost of web design. Because the appeal is subjective, the designer may need many iterations to get where you like to be. As opposed to development where you can see convergence on feature completion and bug fixing.
    By the way the Ruby On Rails podcast just had a podcast (http://bit.ly/9RNZ6u) on http://dribbble.com/ , a site for designers share tidbits about what they are working on.

    Thanks Taylor for the interesting discussion!

  13. @Aaron@Phrasemix Thanks for the kind words! I was nervous to do the show because I’m such a recluse. I’ve been working from home for the past 3 years and wasn’t even sure if I could talk! :p

  14. @Philippe Monnet “He actually seemed to have helped balance the energy flow” Nice :). It’s the full spectrum of energy. Jason is high, Justin is medium, and I’m low.

    Regarding ASP.NET, I forgot to mention is Microsoft Bizspark http://www.bizspark.com . They give startups free software which helps.

    Great tip about http://dribbble.com/! Many of the best designers showing off their latest work/experiments. The quality of designer is high, it’s invite only so the stinkers can’t get in.

  15. RE Swarm pricing: You should release the game as paid first so you make money while you’re working on adoption. It’s much easier to take a paid product and make it free than go the other way around. To add paid to a free product, you need to add a LOT.