The research readout your VP will actually read

Why don't VPs read research readouts?

VPs skip research readouts because most readouts are status updates wearing a research costume. They open with methodology, walk through findings in the order the researcher discovered them, and end — if the reader makes it that far — with "implications" that stop short of asking for anything. The reader is asked to do the researcher's final job: figure out what it all means for the roadmap.

Maya's first draft made every one of these mistakes. Slide 2 explained the study design. Slide 7 introduced participant demographics. The finding that actually mattered — 31 of 42 participants abandoned onboarding at the camera-permission step — appeared on slide 19. Her VP got to slide 4 in the elevator, closed the tab, and asked her in Slack: "so are we shipping or not?"

That question is the readout. Everything else is supporting material. A readout built for a decision-maker answers it in the first two lines and spends the rest of the page earning the answer.

A readout that doesn't name a decision is a status update — and executives archive status updates. Name the decision, make the ask, and rank your evidence. That's the whole trick.

What goes in a one-page research readout?

A one-page research readout has five parts, in this order: the decision as the title, the ask in one sentence, three to five findings ranked by evidence, one line on what would change your mind, and an appendix link. Here is the template, filled in with Maya's study:

TITLE — the decision, not the study
  "Receipt-capture onboarding: ship now or fix the permission step first?"
  (not "Onboarding Research Readout — Q3")

THE ASK — one sentence, decidable, with a deadline and a default
  "Recommend we delay launch one sprint to rework the camera-permission
  step. Need a decision by Friday; default if we don't hear back is to
  ship as planned."

FINDINGS — 3 to 5, ranked by evidence strength, one line each
  1. STRONG — 31 of 42 participants abandoned at the camera-permission
     prompt; 24 of them said they didn't understand why a receipt app
     needed the camera. [quotes →]
  2. STRONG — All 42 who passed the permission step completed capture in
     under 90 seconds; the flow itself works. [quotes →]
  3. MODERATE — 14 participants expected to forward receipts by email
     instead; unclear if this is a blocker or a preference. [quotes →]
  4. WEAK, BUT WATCH IT — One participant photographed a week of receipts
     in one batch, a workflow the design assumes never happens. [session →]

WHAT WOULD CHANGE OUR MIND — one or two lines
  "If the permission drop-off doesn't reproduce with a rewritten prompt
  (retest is 2 days), ship on schedule."

APPENDIX — one link
  Full themes, all quotes, transcripts, methodology, participant screening.

Each part exists to protect the reader's three minutes. The title tells the VP whether this memo is theirs to act on. The ask tells them what a "yes" costs. The ranked findings let them audit exactly as deep as they want to. The change-our-mind line signals that you are reporting evidence, not lobbying. And the appendix link means the memo never has to argue about completeness — it is all there, one click away.

How do you rank findings by evidence strength?

Rank findings by three questions: how many participants showed it, how consistently it appeared, and how directly you observed it. A finding earns strong when it is frequent, consistent, and behavioral — 31 of 42 participants abandoning at the same step is strong because it was watched, not inferred. Moderate covers real patterns with an open interpretation, like Maya's 14 participants who expected email forwarding: the count is solid, the meaning is not. Weak covers anything appearing once or twice, or anything participants said about their future behavior rather than did in the session.

Two rules keep the ranking honest. First, attach the actual numbers — "most participants struggled" invites debate; "31 of 42, quotes linked" ends it. This is also where interview volume quietly matters: evidence rankings get sharper as N — your participant count — grows, which is a practical argument for running 40 parallel AI-moderated interviews instead of 6 calendar-scheduled ones — 5 of 6 is a hunch, 31 of 42 is a finding. Concrete counts with linked quotes are also your best pre-emptive defense against the "that's just anecdotes" objection, because they answer the rigor doubt before anyone voices it.

Second — and this is the format's honest weakness — evidence ranking rewards what is countable. Maya's single batch-photographing participant is her most strategically interesting observation, and the ranking system buries it. The fix is not to inflate its label but to flag it explicitly: weak, but watch it. A VP trusts a memo more, not less, when it shows a finding it refuses to oversell.

How do you write the ask?

Write the ask as one sentence a busy reader can say yes or no to, with a deadline and a default. Maya's ask names the action (delay one sprint), the reason (rework the permission step), the deadline (Friday), and what happens on silence (ship as planned). The default is the part most people skip and the part VPs appreciate most: it converts a stalled decision from a blocked project into a conscious choice someone has to override.

If you cannot write the ask, you are not ready to write the readout. "Share findings with the team" is not an ask. "Decide between A and B by Friday" is. When the ask is the big one — keep building, change course, or kill the feature — that decision has its own structure worth borrowing before you write the memo. And when a study genuinely produces no decision — pure discovery work, for example — say that in the title and ask for the next study instead.

Should you send a memo or present a deck?

Send the memo first; build a deck only if a live decision meeting is already on the calendar. The formats solve different problems:

The deck still has a job: contested, high-stakes calls where the VP wants to interrogate the evidence live. Even then, the memo goes out as the pre-read, so the meeting starts at "finding 3 — is it a blocker?" instead of at slide one. Maya's sprint-delay ask got decided in a Slack thread, memo attached, in four hours. No meeting happened.

What belongs in the appendix?

The appendix holds everything a skeptical reader might want to verify: every theme with its supporting quotes, full transcripts, the discussion guide, participant screening criteria, and counts. The memo makes claims; the appendix makes them checkable. The critical property is that every finding in the memo links directly to its evidence — a reader who doubts finding 1 should reach the 31 abandonment moments in one click, not in a request to your calendar.

This used to be the most expensive part of a readout: assembling quotes per theme across dozens of interviews took days. It is the one part of this workflow that modern tooling has genuinely collapsed — in Sera, synthesis links every theme to its source quotes across all sessions automatically, so Maya's appendix was the study's share link rather than a document she compiled. Whatever tool you use, the standard stands: an evidence label without a working link to the evidence is an opinion with formatting.

One honest caveat about the whole format: a perfect memo cannot rescue research aimed at a decision nobody owns. Before you polish the readout, confirm the person you are sending it to can actually say yes. The memo is a conversion tool for decisions, not a substitute for having one.

How do you start using this template?

Start with the study you most recently finished: write the title as the decision it should have informed, and see whether you can produce the ask sentence and three ranked findings from your existing notes. If the evidence column is hard to fill — you know what participants felt but not how many, or you have themes without quotes — that is a signal about your synthesis, and the thing to fix before your next study, not after it.

The structure is the same whatever the study was about — a signup flow, a value proposition, a pricing page. And if you are a PM running the whole research loop yourself, the PM-specific playbook covers where the readout fits in the week. Run the next study at an N that makes the counts meaningful, keep the evidence linked, and the one-pager mostly writes itself.

Frequently asked questions

How do I present user research findings to executives?

Present research findings to executives as a one-page decision memo, not a slide deck. Name the decision in the title, state your recommendation and ask in the first two lines, rank three to five findings by evidence strength, and link everything else in an appendix. Executives read for the decision; give it to them first.

How long should a research readout be?

One page for the decision-maker, unlimited for the appendix. The memo itself should take under three minutes to read: title, ask, three to five ranked findings, risks, appendix link. Anything the reader might want to verify — quotes, transcripts, methodology, participant counts — lives behind the appendix link, not in the memo.

Should I present research findings as a deck or a document?

Default to a document. A one-page memo is read asynchronously, forces you to commit to an ask, and survives being forwarded. Build a deck only when a live decision meeting is already scheduled — and even then, send the memo as the pre-read so the meeting starts at the decision, not at slide one.

What is a research readout?

A research readout is the summary a team shares after a study — the findings, the evidence behind them, and the recommended action. Readouts traditionally take the form of a presentation deck; this article argues the higher-conversion format for busy stakeholders is a one-page decision memo with a linked appendix.

How many findings should I include in a research report?

Three to five in the memo itself, ranked by evidence strength — enough to justify the ask, few enough that each one is remembered. Every other finding goes in the appendix. If a finding does not support or complicate the decision you are asking for, it belongs in the appendix no matter how interesting it is.