Research That Maps the Whole Citizen, Not Just the Power User

When a PM asks me to research a public-sector service, the first thing I push back on is sample bias. Consumer products can optimize for their most engaged segment. We can't. A government service has to work for the 19-year-old renewing a permit on her phone in two languages and for the 78-year-old who has never used self-service before — both are legally entitled users, not edge cases. So I structure research around the full spread of digital literacy, language, and access, not the average. I recruit across age bands and assistive-technology users deliberately, and I treat WCAG conformance as a research input, not a late audit. The output a PM actually needs from me isn't a deck of quotes — it's a clear map of the core task everyone must complete, separated from the optional depth only some will want. That Layered Complexity model is what keeps a roadmap honest: it tells you what cannot be cut. Research, done this way, de-risks scope decisions before a single screen is designed, and it gives the PM defensible evidence when prioritization gets contested.