A research deck lands in a stakeholder’s inbox.
Forty slides, a methodology section, a satisfaction score, a list of eleven recommendations ranked by an effort-impact matrix.
Nothing happens. It is not that the research was wrong. It is that nobody reading it could tell you, afterwards, what problem it described.
User research storytelling fixes that by borrowing a structure people already know. Films run on setup, conflict and resolution because that shape holds attention and makes consequence legible. Research findings work the same way, and the reason is not decorative: a narrative forces you to say what the problem is, why it matters and what follows.
User Research Storytelling: Turn Findings into Clear Action
1. The setup
Establish who this person is and what they were trying to do. Not a persona with a stock photograph and a fictional biography. A real situation.
“A facilities manager at a school, on a phone, in a corridor between lessons, trying to reorder something she orders every term.”
That single sentence does more work than a demographic profile because it carries constraints with it. Limited time, small screen, interrupted attention, a task she has done before and expects to be quick.
Give the audience the context they need to feel the friction when it arrives. Skip this and everything after it is abstract.
2. The conflict
This is where the finding lives, and it should be uncomfortable to read.
Show what actually happened. Where she stopped. What she tried. What she assumed. What she said out loud while doing it. Direct quotes work harder than paraphrase, because a stakeholder who hears a customer’s actual words in their actual tone stops arguing with the methodology.
Resist the urge to soften. A research report that makes everyone feel reasonably good about the current design has failed at its only job.
Quantify where you honestly can. “Four of five participants” is more useful than “users found it difficult”, and more defensible than a percentage derived from five people.
3. The resolution
Not a list of recommendations. A decision, with its reasoning visible.
The distinction matters. A list invites a stakeholder to pick the cheapest item. A decision framed as “given what we saw, we should do X, and here is what we are accepting as a trade-off” invites them to agree or to argue with the reasoning, both of which are progress.
Be explicit about what you are not proposing and why. Research that recommends everything recommends nothing.
And state what would change your mind. A finding that no evidence could overturn is an opinion.
Why the structure earns its place
It forces completeness. You cannot write act one without knowing the context, or act two without evidence, or act three without a position. A deck of findings lets you skip all three.
It also survives retelling. The person who commissioned the research is rarely the person who decides. A narrative can be repeated accurately by someone who was not in the room; a slide deck cannot.
The GOV.UK Service Manual’s user research guidance covers the practical mechanics of running the sessions that produce this material.
How much research this actually needs
Less than people assume. Five sessions with people resembling your real audience will surface most significant problems, and the same three usually recur.
Perfect recruitment matters less than watching someone who does not work for you attempt a real task while you stay quiet. The staying quiet is the hard part.
This applies to small projects as much as large ones. A five-page site benefits from the same discipline, at a proportionate scale.
Where it connects to the build
Research that names the friction early changes what gets built, which means fewer features constructed and later removed.
That has a delivery consequence worth stating. Where performance and environmental impact both apply, they move together: reducing unnecessary data transfer and processing supports a faster experience while lowering estimated digital carbon, though environmental figures remain modelled rather than directly measured. A feature never built is the lightest feature available.
Our sustainable web design work puts research and structural decisions before visual design for exactly this reason.