User Research Before Roadmaps
User Research Before Roadmaps
A large part of my product work, especially in digital banking, has started with research rather than requirements: market and user research, personas, customer journey maps, prototypes and end-user testing. It can feel slower at the start. In my experience it is what makes the roadmap hold up later.
A persona is a decision tool
Personas are sometimes treated as a slide for the kickoff meeting. They are more useful than that. A good persona helps the team settle arguments about what to build by asking a simple question: would this person actually use it? When the team shares the same picture of the user, prioritisation gets easier.
Map the journey, not just the screens
A customer journey map shows the full path someone takes, including the parts that happen outside the app. In banking, that might be a branch visit, a call to support or a wait for a confirmation message. Many of the real pain points sit in those gaps. A screen-by-screen view will miss them.
Prototype early and test with real users
A prototype puts an idea in front of users while it is still cheap to change. Watching someone try a rough version teaches you more than a long requirements document. It also gives the client and engineering teams something concrete to react to.
Define the MVP by what you need to learn
An MVP is not the smallest set of features you can ship. It is the smallest version that tells you whether you are right. Framing it that way helps keep scope small and makes it clearer what to measure once it is live.
Keep listening after launch
Research does not stop when the product ships. Usage data, NPS and direct feedback show whether the product is working for people. Building incrementally and iteratively means each release is a chance to learn and adjust.
Where AI fits
Building AI products in the Curious PM cohort showed me how much faster parts of this cycle can go, from drafting research summaries to building working prototypes. The judgement about what users need still has to come from talking to them.
Sidhartha skipped presentations and built real AI products.
Sidhartha Pooboni was part of the August 2026 cohort at Curious PM, alongside 24 other talented participants.