I Hate Shipping MVPs

Somewhere down the line, a Minimum Viable Product became something we released to users. A half–finished experience delivered with the promise of new, better features in coming iterations is pushed out to users who are often genuinely excited for what a product can do.

Now, here’s my ‘don’t get me wrong’ caveat: I love the idea of an MVP. I work very quickly when I prototype and it’s awesome to have the impetus and momentum that comes with rapidly realising the visions you had for a product. The core features implemented in a couple of weeks, a usable interface ready to go, it’s usable, right? Let’s release it. Hold up.

Ship It

This fucking phrase gets on my tits. More specifically, the seemingly dogmatic following of it. I’m not sure if it’s been misconstrued and the original shouts from the Silicon Valley offices were actually about already established, beautiful products or whether the call to arms really was to simply just get something done, bootstrapped and out there.

Either way, I’m noticing more and more products being released in varying states of disarray. People point to the first iteration of Twitter and say “Look! It looked like shit! It was just texting!” either ignoring or not realising it was built the equivalent of multiple generations ago, given the pace the web moves at. There are so many tools out there now, documented approaches, that mean we never have to make something that shit and restricted.

I tried this approach myself with Pilcrow and I hated it. I spend a couple of weeks with my head up my arse with this goal of having a releasable product in mind as soon as possible, I cut corners on the design because I was more focussed on getting shit done. Obviously this was my personal experience and could quite well point to me having the wrong mindset when it comes to this approach, but I look at the state of it now and it’s clearly not ready to facilitate the experiences we want it to.

The interface ‘works’, the core ideas are prototyped up, but the whole thing lacks the finesse and ‘in–betweens’ that I like to spend hours, even days, refining. Now a lot of people will say ‘that’s the whole point, do them later’, but maybe I’m a bit silly in that I see those things as integral to the product. They give it personality, a purposeful uniqueness that transcends into lots of subconscious satisfaction that in turn translates into positive affect from the user.

Getting shit in front of a user early is always useful, I’m not condemning that, but I think a huge part of what makes great designers great is their intuition. We’re supposed to be professional designers and here we are delegating product–defining ideas to users, either by straight–up implementing what they ask for, or basing everything that goes in on user response. User need is imperative, but should be dissected and translated into user reward.

I see MVPs as internal tools. They’re something that I create to prove my concepts to myself, my clients, my team. Then I throw them away and start again with the lessons I’ve learned on the real product, the product that I’ll happily put in front of a user outside of the tech world, who isn’t used to using or creating half–done products.

The rush to write production code and the view of a product as a series of features and iterations should come, in my opinion, after the personality and delight is established. I talk a lot about creating ‘environments’ as opposed to interfaces and that’s been a big shift in thinking for me, essentially focussing on the overall emotive qualities of a product first and foremost. It’s easy to get carried away with ‘what should this product do?’ vs. ‘how should this product feel to use?’.


Fail fast, fail often. Someone said that. I can’t remember who, but they were clever. The web is awesome in that we’re actively promoting failure here. There’s not many industries where you can say ‘my initial idea was pretty stupid, now that I look at it; here’s a much better way of doing it.’ Unless you’re House MD or some shit. And this is where MVP falls down a lot for me, I think you need time to reflect on your decisions, time to fail a few times with your concept and understand how it needs to ‘feel’.

Just how ‘designed’ are our products going to be if we’re rushing through the early stages, arguably the most revealing and defining stages of product development, in the name of quickly working features in? Maybe that’s why everything is starting to look the same and 90% of what gets released barely makes it outside of the tech bubble.

Or maybe I’m just cynical and in the way of Real Progress.


If you have any questions, comments, insults, haikus (seriously, I love haikus) or insults, lob them at me on Twitter.