Most guides on research questions stop at “clear, focused, and answerable.” That’s the right starting point, but it’s not where the real value sits. A question can tick every box and still miss the actual problem a business needs solved. The gap between a correct research question and a strategic one is where a lot of research work quietly underperforms. And it’s also where the most valuable work happens, if you know to look for it.
The basics: what makes a research question solid
Before you can make a question strategic, it has to be solid. A solid research question is:
- Clear. Anyone reading it understands exactly what’s being asked, without second guessing.
- Focused. It targets a specific user group, user behavior, or business problem — ideally one with a measurable stake, like revenue, retention, or cost, not a general area of curiosity.
- Researchable. You can actually find evidence and insights for it, through data, interviews, or existing sources.
- Relevant. It connects to a real business goal (or an OKR or KPI).
- Feasible. You can answer it within team’s time, access, and resource constraints.
If a question fails any one of these, it’s not that it’s a bad idea. It’s just not ready yet and needs some adjusting.
How to get there
The path from a vague interest to a workable question is fairly consistent, whatever you’re researching:
- Start broad. Pick an area the team cares about and that’s grounded in something real, for example a recurring complaint in support tickets, a metric that’s changing and nobody knows why. “Onboarding isn’t working” is a fine place to start, not an end point.
- Narrow it down. Broad topics are unresearchable by definition. Narrow the variables: which segment, which part of the journey, which behavior, over what time frame. “Onboarding isn’t working” becomes “Why do users on the free plan drop off before completing setup?”
- Get clear on your objective. Are you trying to explain a drop-off, compare two user groups, or understand why people churn? Your objective shapes the wording and structure of the question, and it’s also the moment to check who the answer is actually for.
- Turn it into an actual question. Open questions like “How does…”, “What factors influence…”, “To what extent does…”, or “Why do…” push you toward analysis instead of a status update. “Improve onboarding” isn’t a question yet; “What causes users to drop off between signup and first value?” is.
- Stress-test it. Can you get to this with the users, data, or access you actually have? Is it too broad to point to a decision, or so narrow it only describes one edge case that won’t affect the roadmap?
- Revise until it’s solid. Compare “Why do customers churn?” (too broad to act on) with “Why do customers on the mid-tier plan churn within their first 60 days?” (focused enough to design a study around). The difference is almost always specificity: a named segment, a named moment in the journey, a named behavior.
This process will get you a solid question. It won’t necessarily get you a strategic one. That’s a different filter, and it’s the one that’s easiest to miss.
The second filter: is it strategic?
A good enough question becomes strategic when it does more than answer something accurately. It builds a deep understanding of users, shapes product value, and designs solutions that actually work. Without answering such a question, big bets don’t land. A strategic question should:
- Fill a real gap. Answer something nobody in the organization currently knows, rather than confirming what’s already assumed.
Usually there’s a lot of knowledge within an organization, however it might be scattered across teams, in need of updating, poorly structured, or missing a shared understanding. - Have practical weight. The answer should be able to impact a decision, not just add color to a slide. A useful test: what happens to the business if this research doesn’t get done?
- Be arguable, not just descriptive. Strategic questions should produce answers that challenge the status quo: existing assumptions, knowledge, or narratives.
This is where most research goes wrong: not in getting the wording right, but in never questioning whether the strategic question was asked to begin with.
The layer beneath the brief
Here’s the part that doesn’t show up in most guides: the question you’re handed is rarely the question worth answering.
I noticed this pattern looking back at a decade of working on strategic research projects. Teams would come to me with a tactical ask: “update our personas,” “fix this backlog,”, “find user pain points”, “segment our users.” But the real problem was almost always sitting one layer underneath the brief. A layer too strategic for a design task, too granular for the C-suite to touch, and too time-consuming for most PMs to own. That layer is where the best research work happens: turning mess into clarity, and clarity into a growth strategy a team can actually act on.
Most briefs live one level below where the real leverage is. Not because anyone’s asking the wrong question on purpose, but because most briefs are simpler than the problems they’re describing. Spotting that gap is a distinct skill from writing a clean research question, and it’s the one that actually determines whether the research changes anything.
Making your question more strategic
A few habits make this a repeatable pattern rather than a lucky reframe:
- Treat the brief as a starting point, not the destination. Ask what problem the brief is a symptom of, not just what it’s literally requesting.
- Ask who else would care about the answer. If the honest answer is “only the person who wrote the brief,” you’re probably still one level too tactical.
- Ask what the business will do with the answer. A question that only produces a better description hasn’t earned its strategic label yet. A strategic question needs to point toward a decision.
- Notice when you’re being asked to solve something too small for the problem underneath it. A useful check: what would happen to the business if this research simply didn’t happen? If the honest answer is “not much,” you’re probably still one level too tactical. That’s usually the exact moment worth pausing on.
A good research question is clear, focused, and answerable. A strategic one asks whether you’re even solving the right problem before you get there. The craft isn’t just in writing the question well — it’s in noticing, before you commit, that there’s another one sitting a layer beneath it.
