I read something the other day that has really stuck with me, an analogy given by Eric Ries in his book The Lean Startup.
To summarize, companies usually want to plan everything before they get started. Lots of planning meetings, documents and user studies to create a great spec document so you can all go off and build your product, no worries. This mind set is a bit like building a rocket. Everything is pre planned, down to the finest detail, months before launch and then it all just happens at once. 3 2 1 Lift off!
The problem with this approach is that the smallest problem or error in the calculations could cause a catastrophic failure and the launch is aborted or worse.
Eric puts forth the idea that building a product or company should be similar to driving a car instead. While you may have a basic route planned out before you set off on your way, the actual actions of steering the car are not. You just do it when it comes to it. This allows you to adjust your course as you continue on your journey with ease. Road works? Take a different way. Low on fuel? Stop for gas before you are stranded on a highway somewhere.
With this kind of thinking applied to building products, you can see problems as they arise and adjust your course to deal with any issues faced.
This is why I like the concept of validated learning.
By validating your hypotheses as quickly and simply as possible, you have the ability to adjust your course and even change route (pivot the product) before you find yourself stranded on that highway with no money and a product no one wants.
The process should look a little like this.
Identify a problem.
Create a hypotheses on how you might fix this problem.
Create the most minimum viable product (MVP) to prove if the hypotheses is worth pursuing or not. This could be as simple as a few screens in an InVision project or a fake landing page to gauge potential interest in a feature.
If it doesn’t work, you haven’t wasted to much time/money on it and you can start working on a new hypotheses. You have learnt something valuable even if it fails. You are experimenting with your solutions.
If it does work, continue to test as it grows into a product to ensure your are steering it in the right direction.
Never stop testing and re-adjusting, because a hypotheses can become invalid at any time, needing a quick pivot to make sure your product will have the best chance at success. Hopefully it’s not that serious and you can just change course a little to address the issues and be back on track, but always be validating the work you are doing.
The earlier you can see a potential solution is not going as planned, the easier it is to try a new approach. The longer you wait, the more invested you become in the solution, making it harder to give up.
by Alex Knight
UX / UI Designer, Tokyo, Japan.