MVP is hard

June 25, 2014 / Start-up, AvidTap, Kash, AvidRetail

People in the startup world probably have all heard of the term Minimal Viable Product (MVP). On the other hand, people in established companies might not know the term at all. Even if they know of the term, they might not have a good feel for what it’s all about. I actually think it takes a while to get a feel for it. I think it’s something people continually get better at and have to keep getting better at. I say this from experience based on the number of mistakes I made so far. Perhaps it’s because it’s counter-intuitive in the big company world, which is sometimes the only thing some people are in.

When Geoff, Kaz and I started the company we took a bunch of time on building our first product, which was a mobile loyalty app. We officially started in Nov 2012 and we did not get into a store until March 2013. I still shake my head at the time wasted. Even though we built a bunch of cool technology, we weren’t focusing on the right things. We experimented with sound based data transfer technique, we built a NFC reader that interfaced with Android Beam, and we even built an app for BlackBerry. All of which costed us time but were low impact. Just as studying a subject in school doesn’t necessarily mean you know how to apply it. Same thing here.

During HyperDrive when we created AvidRegister, we got way better and way more focused. First of all, we sold it before we had it completely built. We only had some initial screens, just barely enough to give a glimpse of what it might look like. That was our MVP which helped us have Kaz go around selling it and see retailers’s reactions. Let’s say that MVP was 1 week of work. We then took 1 month to build it and launched live into our first store. So I definitely felt that we’ve gotten a better feel of it. We got better at doing things that matter.

With our P2P product, we again did it from start to finish in 1 month. In that timeframe, we’ve put the product in people’s hands, talked to potential users and talked to actual users face-to-face, and even went through the necessary legal procedure (although that was perhaps not necessary). Even then, I still wished we had done things way faster. For MVP you really have to be ruthless about exactly what you’re trying to answer and trim everything unnecessary, then just when you thought you have the bare minimum, trim some more.

For what we’re working on now, we’ve gotten the MVP out in 1 week. We originally thought we’d take 2 weeks, but a critical examination showed us that we could get certain questions answered faster and another set of questions answered subsequently. We got better again. Talking to other founders, I now can spot when people are spending too much time on things that don’t matter. So MVP is definitely not easy, and if all you know about it is reading up or hearing someone talk about it, then you probably don’t get it still.

Not sure what’s the best way for this kind of experience-based knowledge transfer. Perhaps try to be super aggressive to the point of feeling like the timeframe is impossible, then examine if there really aren’t any pieces to delay. I think being able to talk to other founders probably helps too. A second opinion is quite helpful in this.