I just got an idea
I had several project ideas in my life, but never thought of writing up the process leading to the making of the MVP, or even the company building.
This time, I decided even if the idea may not work, I won’t let it fall into darkness, and share the knowledge I gathered throughout the conception. I’ll try to stay pragmatic so you can read up the big lessons I’ve learnt; may this help you avoid doing the mistakes I did.
The seed of Alfred
I travel a lot, and because of that I had to find alternatives to the usual shit bank you find in France. One of the best for currency exchange and in term of UX is Revolut, from the UK. Long story short, after years of using it I decided to go for the “metal” offer, which includes a “conciergerie” service.
First of all, what the heck is that? To make it simple, people who have elite / black credit cards usually get lots of bonuses with it, this often includes a “conciergerie” private line.
It’s basically a service you can call up which helps you book a hotel, a restaurant, a private jet, or whatever you desire. It’s classy and personal. They find you the good deals, and you don’t have to look by yourself.
This was originally thought through for relatively rich people, or let’s say upper middle class individuals. In its most basic form, it’s just delegating the little stuff to someone else, which’s what all rich people do with their house, finances, etc. Because when you have money, your time is more precious or something.
Revolut is very modern, and usually original in its way of dealing with stuff, so they included this service for much lower price than usual, and in a “chat” form. I thought great! I can get drunk with my friends, open up my phone and write down a message to Revolut to book whatever place I want to go to tomorrow. No need to call anymore. Who has time to call anyway?
Except they did it all wrong. Their service basically take days to answer, you can’t interact properly with the guy because it isn’t really a chat, but a slow-ass ticket service, and all they do is look through Google. No “taking your money directly to do the stuff” despite being a bank. They can’t do shit, for real. Worse than that, when the ticket isn’t satisfying, you’ve to send a new one, literally. It has nothing to do with “real” conciergerie service where you call up and they do the stuff.
But the idea beneath this was too good. Without knowing how they orchestrated it, I was genuinely excited to use it and I crawled under requests ideas. After a few attempts, I just gave up because of their bad execution.
The problem is that, after and for a few days, I really wanted to use something like what Revolut was supposed to propose, that’s when the idea of Alfred emerged.
I wanted to make Alfred the way I wanted Revolut to be. Simple, straight forward, reactive and efficient.
Do I need to choose a plumber in a specific price range for 8am tomorrow? Just ask Alfred and he dispatches it all. My administration needs something sent by mail but I don’t want to move my ass? Alfred has all my details already — because it bulks up progressive knowledge about me — he can fill it and send it within a day with my approval. You can delegate the stuff from all the aspects of your life, depending your budget and needs.
Now, that’s an idea that can go far.
This time, it should be different
I made a ton of projects in my life, and I can proudly say 80% were never even commercially released, and less than 5% were decently successful. In term of pure tech projects, none were actually successful. I’ve never made a real company out of a pure tech project.
I decided to change this. I knew I “got something” with this idea, but I wanted to do it better this time, maybe there’s something wrong in the way I plan the execution of my projects.
As a developer, I usually work on things by going through the idea itself extensively, thinking how I want it materialized — more or less — and go straight up to code the pages with an idea of the user interface to compose, the global theme / atmosphere, and a specific tech stack I think is adapted for the project.
I would then “improve” the graphic design empirically by adapting the code, as I originally thought I had a large enough experience in term of UX / UI.
For me, that was the definition of a good MVP. After all, on your first MVP you do one big mistake: work too hard, and never release it — which I did a few times.
I learnt much after there’s another big mistake: make it too minimal, and/or release a bad version of it.
Even though I was now systematically releasing my projects, they were all so-so at a few levels. You know, the kind of project which is functional, got all the basic features, but doesn’t seem so well done at the end? That’s most of my projects. The idea can be great, but the execution should be as well.
This strategy of going fast can be successful, I’m sure of it, if the idea is good enough and highly addictive, you can then think “yeah once I got the user base I get a team and we go crazy on it”. Well, if you find the end result average, so will your users, consciously or not.
I’m not talking about code quality here. But the execution overall. The graphics, illustrations, storyline, spirit / atmosphere, UX, CX … Everything but the code. In short, the vision.
