The Three Week Startup – Keith Perhac – MicroConf 2015

0:00[Music] [Music]

0:14hi I'm Keith fer heck I flew in from Japan you might know me from the calumnious podcast I also do some it evergreen and today I'm going to talk about the new sass that we launched last month called seg metrics I'm going to talk about building a sass in one week so ok not clicking so I'm going to use that that's not working either there we go hi I'm Keith / hack flew here from Japan boy and my arms tired got to do that one I talked really fast I also only have 12 minutes so I'm just going to fly through these as fast as possible notes will be available

0:47afterwards so I run a technical marketing consultancy we're about eight people strong we do conversion rate optimization helping Sasa's and internet marketers improve their sales funnel I'm also the co-founder of summit evergreen which you may have heard of from Mike as well as seg metrics when I'm going to talk about today I'm also the co-host on the Cal's Mia's podcast but I already said all right so here's the challenge we had about two weeks in between some client work and we were like okay what can we do with this downtime we've had some great ideas flying around let's see if we can launch a product in 50 hours

1:17ok so I have 12 minutes three points finding product market fit insane pragmatism and think of marketing from the start and I probably won't have time to talk about that disclaimer this is probably not the best way to launch assess it's maybe not even a good way to launch assess the entire reason I'm talking about this today is that there are many people out there that have been working on a product for a year or two years they just got to do one more thing before they ship and I just want to show you that it is possible to get that thing out the door and and start doing

1:46it so the first thing is to find an idea so best way to find an idea is of course to scratch an itch now you always hear this everyone wants to do a time tracking app or something with guid or some bug tracking app good ideas out ideas there's a lot of them but there's really two ways that you find a niche that you want to scratch they're either tools that you want internal that would be like time tracking or something or tools that you need for work which are external the things that people have paid you to make in the past or things that people pay you to do that can be

2:16automated and you need to like patrick said make sure it's something you're interested in because you're going to be doing this for a long time great way to vet your ideas are people already paying you to do it manually so that's where we came in seg metrics we do a lot of Infusionsoft consulting we pull a lot of leads and find lead values for different segments and different lead funnels and it's really a pain in the butt it requires a lot of exports data pivot tables and Excel making pretty PDF charts and everything it takes us about 10 hours to do one report and we're like this can be automated and

2:51if we want it to be automated there are probably people who want to automate it as well now the second thing is once you have an idea you need absolute insane pragmatism so the number one rule of MVPs is that scope creep kills so I've worked with big companies like Toyota I've worked with small companies that are one man I've worked with people with companies that are about 10 people especially bootstrappers you have a limited amount of time a limited amount of money and a little limited amount of effort that you need to be able to get something out the door so if you're focusing on the absolute perfect thing

3:25that's going to solve all your customers problems forever you're never going to launch as Voltaire said perfect is the enemy of good so here's something specific examples that we had as we were building this out we tried to be very dramatic Prague Matic about the development this is one thing that a developer had said to me hey I think I could spend a day and make this MySQL query faster all right it was already pretty fast but we can make it faster that's 8 hours of a 50 hour time budget I don't care should we support multiple crms to start no of course not we don't even know if

3:57anyone's going to buy this for one serum when I support multiple ones might as well have it in the database anyways but not build it if it is not in the core proposition you need to get the hell rid of it all right did the core proposition change maybe your your users gave you some bad feedback good thing you didn't waste all that time on technical debt you spent a week building something you find out someone note no one wants you either have to throw it away which let's be honest no developer ever wants to throw away a week's worth of code or you have to reuse it and leave it in this

4:25technical garbage right you need to be merciless with which features you launch with if it's not providing core value to your customers you do not want it in there so here's some famous examples Basecamp did not have billing code for the first month they said we get customers in we have 30 days to figure that out no third-party apps on the iPhone when it launched no account limits pretty much every SAS ever made I think bingo karma part my render might still not have it I know that summit evergreen still does not have it and seg metrics still does not have it so some examples for seg metrics no account

4:57limits no queues for data gathering no automated data gathering you had to click a link to do that 30-second load time for the graphs number of people cared zero number of complaints of course zero number of people noticed one and that was me so it's only for me that I wanted to have all these things in and that's a bad proposition right until you find a fit the less technical debt you have the better type feedback is key so you go from an idea feasibility concept feedback from real people real customers who are willing to pay you money so this is the first feedback that we gave three

5:34hours in we sent it to one of our beta customers this was all generated from their actual data and it's number of leads number of buyers conversion rate number sales number amount of revenue etc we built down all the core metrics that we give people and we said is this useful if you were to able to look at this once a week for any segment you had it any lead source you had is that useful they said hell's yes all right so we knew we had a good thing to go for one thing that's very important and what this previous slide shows a feature is not a feature if you cannot show it to

6:02your users if we could not show those numbers repeatedly to a customer it doesn't matter that they're there it doesn't matter how cool the internals of our app is if the customers who need to be giving us feedback cannot see it and leverage it themselves so I want to talk about one thing that a lot of developers and a lot of people really don't get is that it's all about building blocks you build the simple crud first you build the core features available in the browser it doesn't matter that it looks like poo okay as long as someone can see it someone can leverage it it is

6:34important and as developers we don't want to do that we want to do the fun stuff we want to do the hard stuff but that's not helping your customers really cool picture strong foundation lets you show-off sooner showing off soon or get you feedback quicker quicker feedback you improve faster which brings us back to the idea feasibility concept feedback booyah alright don't try to be too slick this is really important just like we like making we like working on the hard problems we're entrepreneurs we're developers we've like solving problems that's not going to help you right here right it's more important to get great feedback on your core offer than have

7:12something super polished complicated and slick and I think this really important makes it harder to pivot and the reason is it's because mentally you're attached to that great idea and technically you have a lot of code that you don't want to throw away because now you found out that your core proposition is wrong also when we were building some whatever grain we did not follow any of these tenets and the UI for someone evergreen I love it it's this three pane thing it's all ajaxy and awesome it takes us so long to make an update we want to add a page we have to test so many things as

7:43a pain in the butt I wish I had made it simpler you have much more time after you launch than before this is something that people really forget I forget as well everyone thinks that it's the goal to launch after that you're done you're set you're gonna sit on the Bahamas I have a million dollars it's the opposite it's like getting married there's life you were like marriage is the gold no marriage is the first step after that is the long haul that's when you've got to make it work alright let's talk about marketing the iPad actually turned off so I don't know how long I've been

8:14talking so if you build it they will not come this is very classic and a lot of SAS places that they're talking about this kind of stuff just because you build it no one cares most important thing is to find out where your market hangs out if you don't know where your market hangs out a great way to find that market is to build an integration building an integration gets you access to that market stripe is a good example with bare metrics be stunning lots of lots of things integrate now with stripe Infusionsoft which is the one that we chose because we do a lot of Infusionsoft and

8:46Grayson's ontraport drip Gumroad all these places if you integrate with them you now have a ready-made market that not only you can talk to and specifically target you can also do Facebook ads and stuff to that target market being having an integration doesn't mean that you're stuck there forever you can always pivot away later famously I can't remember a B Airbnb started out doing only Craigslist they pivoted away you can always change later having that instant market already made for you is a great win wish I had so there's a bunch of things I wish I had done I wish I had built an email list to start out with I

9:25wish I had gone on beta list and started building an email list I wish I had built lead magnets and some sort of information to get people in and that's what we're doing now it would have been nice to do it before but we this was more a practice of is it possible we had done this all was somewhat evergreen we had done the full okay let's build up a list let's do the marketing let's do all this stuff and then launch it this is can we get this idea out the door all right I think I've actually talked over the time but the iPad has not gone off

9:55so I'm just going to go out of time pass on to the next person you can find some information as well as we put together some extra specials on the little USB key that you have as well as seg metrics tayo hi micro comp thank you very much everyone [Applause]