Over the years I’ve written a lot about starting with solutions. Like when we needed to improve publishing times on GOV.‌UK, when we thought structured data might help with ‘personalisation’ and early ideas for what ‘personalisation’ features might be, plus the time we saw an opportunity to make GOV.‌UK Pay available for call centre staff.

Realistically these weren’t final solutions but early ideas – hypothetical solutions – that we could test out and refine (or throw away) based on what we learned.

I like starting with hypothetical solutions and using those to learn more about the problem because they give you a starting point. This comes from using Lean product methods in my work a lot, thinking big and starting small.

But two people have written posts in the last couple of days that really nail why, although it gets a bad rep, starting with solutions isn’t always bad. And one person shared a framework to help.

Andrew wrote about rather than avoiding stakeholder ideas, leaning into them and making them part of your process. And Kuba wrote about how some problems are high-certainty, and can be solved with commodities; others are completely uncharted.

Vicky has also shared a diagram to help work out how much certainty there might be, so you can move forward to the appropriate stage.

The key is realising the link between problems and solutions, how learning more about one informs the other and vice versa. It’s why we need to talk more about understanding problems and exploring solutions as a virtuous loop, not sequential phases. Alpha is discovery.

Sure, there will be times when starting with research – because we know nothing or the domain is really risky – is the right thing to do. But we’re not always starting from zero.