How to Build an MVP

How to Build an MVP

by Miguel Tavares -

It’s a little known fact that around 90% of Startups fail- and there’s no end of articles and books on how to avoid this. Everything from a lack of focus to just the wrong timing can mean your product or app doesn’t take off.

Granted, a lot of these factors are out of your control. But there’s one thing you can control - and that’s your MVP or prototype.

Now again, there are plenty of MVP frameworks out there and lots of techie advice on the best and worst applications to use in your MVP design.

But there are some simple and basic principles you can follow that will help keep you on the straight and narrow.

In fact, we’ve worked with dozens of startups over the years and built a variety of different apps and MVPs - so we’ve compiled our top tips on how to build an MVP into this short blog.

1) Don’t build the MVP first!

Now this may seem a bit counterintuitive but before you plough lots of time and money into building an MVP, build something a bit more basic first.

This could be a clickable prototype in a low-code application such as Bubble or even InVision, or it could be a landing page that you create as part of a simple marketing campaign, or it could just be a slide deck with a walk-through of your idea.

The most important thing here is that you validate the concept and make sure you are solving a) a problem and b) the right problem. It’s crucial that you get that ‘product market fit’ before you start building anything.

2) Validate, Validate, Validate

So you’ve done your initial validation and are moving ahead with the MVP build - congratulations!

But your validation doesn’t stop there.

In fact, it’s useful to think of validation in 3 main areas:

  • Tech validation - so a validation of the problem you’re trying to solve and the solution you have for it.
  • Business validation - what are your business goals and objectives with the MVP? Do you want it to be producing revenue from Day 1 or is it purely something you’re building to demo to investors and the like?
  • Pricing validation - there’s no point just validating a free version of your tech unless you intend on keeping it free. You’ve got to answer the difficult question of if your customers are willing to pay for the product - or what features they are willing to pay for if you want to get them in at the freemium level.

Look out for an upcoming blog we’ll be doing on the whole topic of Continuous Validation!

3) Focus the problem

If you follow principle #2 then inevitably you will come up against any number of problems that you could be solving for your customer. And that’s great - not only does it show engagement but it shows you’ve landed on the right area and problem to solve!

It’s therefore tempting to get carried away here with building lots of bells and whistles into your MVP.


Don’t do it! The most important thing at this stage - for your customers, your investors, and your own sanity - is to focus on a few key things and to build them simply and easily.

A great example here is if you’re looking to include machine learning into your product. Well for the MVP, don’t build the model - in fact, don’t even build any sort of automation! Do it manually - think Wizard of Oz and have a person ‘behind the curtain’. This will not only help keep your costs down, but it will also once again help you validate the problem and perhaps even come up with a different and more cost-effective way of solving it.

And there you have it! Our top 3 tips for building an MVP in 2019!

If you’d like some further advice on how to build an MVP, or even if you just have an idea you’d like some input on, then why not sign up to one of our free Office Hour sessions. We run these regularly in Shoreditch, London and would love to see you there!

#mvp #startups

Miguel Tavares

More posts

Share This Post

Right Talent, Right Time

Turnkey technical teams and solo specialists
when you need them for greatest impact.

Work with us