For several years after creating and then leaving Ansible, I helped out with the Senior Design project course in Computer Science at my alma mater. From there, as I had seen on forums like Hacker News and Reddit, many students used the phrase “Minimal Viable Product” a lot.
Each time, I tried to not have some sort of conditioned response, because obviously, it wasn’t their fault for believing in this strategy, it’s just one of those common “startup religion” phrases that everyone throws around and doesn’t really know how to put in context. In short, they were using a product management strategy to describe software prototypes.
I see a lot of people glamorizing startups and not realizing they are a ton of work and they maybe don’t have the knowledge to have their own company yet. Believing in MVPs has a lot to do with that. Anyway, the phrase makes me twitchy.
What is there to be careful of MVP as a strategy? Well, to start, I don’t think MVP works. The risk is you’re going to build something and nobody will come (or stay), and then you’ll run out of money.
This view about “MVPs” just being first releases makes people believe, I think that startups are easy, and it goes back to what I think was a Steve Blank definition - “A startup is an organization formed to search for a repeatable and scalable business model”. That might be ok, and I’m ok with maybe figuring out pricing as you go, or trying to figure out whether support and services is effective. But no - who would give you money if you didn’t know what product you were making? The product should be in your DNA. This definition only works if we think of the business model as the spreadsheet-equivalent that tracks income, expenses, and runway as you change what you charge for and how much.
The whole MVP process presumes you don’t know what you are making for a product, and that is death. You have no soul as a company at that point, you’re just trying to make money with other people, rather than help people with a problem that you know. You may think you have an idea, but it’s not good enough yet.
Your roadmap? Know what you need to do. If you have a bunch of product managers that don’t know, get new ones with more experience and better instincts. There are a reason they are few of these at companies - they can do more damage to a company and end it faster than a bad CEO.
The original idea about MVP has a lot of parallels to the mostly imaginary “on site customer” that Extreme Programming and the Agile Manifesto were designed around and require (it maybe works for consultants?) that later morphed into the heavily structured mythology that is Scrum. Anyway, MVP? Asking your customer to tell you what products you should make?
Often customers don’t know, they say yes to everything you propose, or they don’t know what to prioritize. They give requirements they don’t really have. (Given, I love customers! I do! I would ask them what they want, but I’d ask dozens and average ideas together if I could, and then couple that with where I want the market to be, because maybe they don’t see the same possibilities! Know the market, survey your customers, but know the market because you are having a lot of conversations)
Don’t have a prospective customer audience? You definitely can’t play the MVP game by faking it. Declaring someone a “product owner” means this person must be a domain expert and essentially your head of product. If this person isn’t? Possible failure incoming. If you don’t know how to make people try your thing out, you can’t get feedback to evolve the MVP.
Anyway, MVP's… what part of “Minimal”, “Viable”, or “Product” is the emphasis on? Most people would say Minimal. But no! In a startup, everything is on a timer and this should be lesson #1. With time, you’ll be too fat to move, and some other tech fashion trend will have replaced you or you will have taken in so much money you can’t make an exit that isn’t a fire-sale. The emphasis should be on viable and product, and getting there ASAP.
An example of too much minimalism would be failing to implement UI pagination and your UI falling over with more than 20 objects, because you didn’t think about the use case. Maybe you haven’t thought about scale-out or upgrades or you can install on only one operating system. A serious buyer can sniff out this incompleteness and won’t take you seriously, and only serious domain experience can fill in these gaps. You can’t expect your buyer to be like “hey, nice try with this prototype but I can tell this won’t scale to more than 20 objects”. They are looking to you to be the expert.
An example of non-viability or non-productness is if you want to have a business, but you’re not quite positive what you should build or sell. This is no good as the product and the desire to help your customer, rather than to just “have a business”, should be your soul.
Time is short, funding is short (especially if you are running on your own money), and you don’t want to waste time.
Your company shouldn’t be in search of a business model, you should know what you need to build. Wait until you see the exact right thing, because you are about to put too much of your lifeforce into it, and if you don’t, you have a much harder road ahead. You want to know it will work, and this requires cultivated instinct. When the time is right and you know enough, that’s when you decide to have a product.
When I see students right out of school trying to start their own company, I really wish they would wait, try to find a problem that they know exists and a market they think they can really reach, and then maybe try to solve that later - not only will they get more domain experience, but they’ll see how 5 other companies have built something and know all the things that they’ve done wrong, and it will be a lot easier.
Another thing online startup bro-culture tends to do (or did, historically) is celebrate failure. Don’t be mad at failure though, but all that takes time and tons of energy. You want to celebrate choosing your battles. Wait and try to pick ideas that don’t fail. Be super selective.
A large product “pivot” should be considered a failure, it’s just a failure that your investors allowed you to keep your jobs over. If you pivot your marketing too much (less bad?), you won’t let your customers understand what you do, and you will have not given things enough time to stick. You want to build successful products and this requires knowing a space and doing them right the first time.
Yes, start small, and have a roadmap, but know what that first release criteria needs to be, and nail it completely. Get some paying customers.
This is why I don’t try to offer product strategy consulting anymore - if you don’t know what your product should be, it’s not going to work, and I probably can’t fix you.
Lots of people can still use a lot of help communicating what it’s their head, but if they lack the domain experience, someone with less domain experience can’t help them out and teach them what that product should be. And to me, that’s why MVP doesn’t work. It’s for people that like the idea of a company.
If you’re just using “MVP” as shorthand for your very first public release, maybe don’t call it that, because it stresses the wrong things, and can make you build the wrong things.
Got that idea for a product and the featureset nailed?
In upcoming posts I’m going to talk about the mindset of how to navigate a startup through various minefields (including competitors and the market) and also some of the downsides that that mindset seems to have long-term.