{"id":"res-pm-ux-findings-exec-bullets","audience":"product-managers","category":"research","technique":"grounding","title":"Turn a usability study into three counted findings for execs","free":false,"task":"Your VP wants the usability study results before Friday's roadmap review. You want at most 3 bullets in plain text, each one finding with how many of the 8 participants it applied to (like \"5 of 8\"), only findings seen in at least 3 participants. Use only the study notes and don't add recommendations.","materialLabel":"Usability study notes (attached to your prompt automatically)","material":"USABILITY STUDY - New checkout flow (8 participants, Oct 2026)\n\nP1: Couldn't find the promo code field. Completed checkout in 3:10.\nP2: Missed the promo code field. Liked the saved address feature.\nP3: Completed fast. Asked if Apple Pay was available (it's not).\nP4: Missed the promo code field. Confused by \"Express\" vs \"Standard\" shipping labels.\nP5: Confused by shipping labels. Liked saved address.\nP6: Couldn't find the promo code field. Abandoned at the payment step.\nP7: Confused by shipping labels. Liked saved address.\nP8: Said the button color was \"kind of ugly.\"\n\nResearcher note: P6 abandoning is a big deal, we should redesign payment.","rules":[{"type":"maxBullets","max":3},{"type":"plainText"}],"mistakes":[{"title":"No counts","why":"The AI writes 'several users' or 'most users', which execs can't weigh.","fix":"Say: \"Give each finding as N of 8.\"","checks":["r0"]},{"title":"Promoting one-offs","why":"A dramatic single moment, like one abandonment, can get treated as a finding.","fix":"Add: \"Only findings seen in at least 3 participants.\"","checks":["r1"]},{"title":"Adding recommendations","why":"The researcher's note pushes the AI to suggest a redesign.","fix":"Say: \"Findings only, no recommendations.\"","checks":["r2"]},{"title":"Not capping the list","why":"The AI lists everything it finds, so the few points that matter get buried.","fix":"Say: \"at most 3 bullets.\"","checks":["maxBullets"]},{"title":"Getting markdown back","why":"AI defaults to bold labels and headings, which show up as stray symbols in a Slack message to your VP.","fix":"Ask for \"plain text, no bold, no headings.\"","checks":["plainText"]}],"promptHints":[{"label":"Asks for counts","patterns":["of 8","how many","count","out of"]},{"label":"Sets a threshold","patterns":["at least (3|three)","3 or more","three or more"]},{"label":"Rules out recommendations","patterns":["no recommend","findings only","(don't|do not) (recommend|suggest)"]},{"label":"Caps the list","patterns":["\\b(3|three)\\b.{0,15}bullet","at most"]},{"label":"Asks for plain text","patterns":["plain","no bold","no markdown","no formatting","no headings"]}],"hint":"Execs need to know how common a problem is. Ask the AI for counts, set a bar for what counts as a finding, and keep recommendations out.","rubricLabels":["The findings are among: missed or couldn't find the promo code field (4 of 8: P1, P2, P4, P6), confused by shipping labels (3 of 8: P4, P5, P7), liked the saved address feature (3 of 8: P2, P5, P7), with correct counts","No bullet is a one-off (Apple Pay, button color, or P6's abandonment presented as a pattern)","Contains only the bullets: no intro, no recommendations (such as redesigning payment), no closing line"]}