{"id":"res-mkt-press-release-survey-stats","audience":"marketing","category":"fact-checking","technique":"verification","title":"Label each press release stat against the survey data","free":false,"task":"Your PR agency drafted a press release around your pet-owner survey, and you need to approve the numbers. You want exactly 5 numbered lines, one per stat in the release, labeled Matches, Wrong, or Not in survey, with the survey figure after any Wrong label. Use only the survey data, plain text, and nothing else.","materialLabel":"Survey results and the press release stats (attached to your prompt automatically)","material":"PAWPATH PET OWNER SURVEY 2026 - Results\nRespondents: 1,204 US dog and cat owners, surveyed Sept 8-19, 2026\n- Spend more on pet food than 2 years ago: 68%\n- Have bought a pet product because of a social media post: 47%\n- Feel guilty leaving their pet alone during work: 57%\n- Use a pet camera: 22%\n- Would pay more for food made in the US: 49%\n\n-----\nPRESS RELEASE DRAFT - stats used\n1. \"A survey of more than 1,200 pet owners...\"\n2. \"Two out of three owners now spend more on food than two years ago.\"\n3. \"Nearly half have bought a product because of a social post.\"\n4. \"One in three uses a pet camera to check in during the day.\"\n5. \"Pet owners spend an average of $1,100 a year on their pets.\"","rules":[{"type":"numberedItems","count":5},{"type":"plainText"}],"mistakes":[{"title":"No fixed labels","why":"The AI comments on each line in prose, which is hard to approve or reject quickly.","fix":"Give them: \"Label each Matches, Wrong, or Not in survey.\"","checks":["r2"]},{"title":"Calling missing stats wrong","why":"The $1,100 figure isn't in the survey, and the AI may judge it by what it knows about pet spending.","fix":"Add: \"If the survey doesn't include it, label it Not in survey.\"","checks":["r1"]},{"title":"Not converting fractions","why":"\"One in three\" can slip past unless someone compares it to 22%.","fix":"Say: \"Convert phrases like 'one in three' to percentages before comparing.\"","checks":["r1","r0"]},{"title":"Not saying how many","why":"The AI decides the length itself and usually returns more lines, one per stat than you can use.","fix":"Be exact: \"exactly 5 numbered lines, one per stat.\"","checks":["numberedItems"]},{"title":"Getting markdown back","why":"AI defaults to bold labels and headings, which show up as stray symbols in an approval email.","fix":"Ask for \"plain text, no bold, no headings.\"","checks":["plainText"]}],"promptHints":[{"label":"Gives exact labels","patterns":["matches","not in (the )?survey","label"]},{"label":"Names the source","patterns":["only.{0,30}survey","survey.{0,20}(only|source|truth)"]},{"label":"Checks number phrases","patterns":["convert","percent","one in three","phrases"]},{"label":"Says exactly how many","patterns":["\\b(5|five)\\b.{0,25}(line|stat)","exactly \\d+"]},{"label":"Asks for plain text","patterns":["plain","no bold","no markdown","no formatting","no headings"]}],"hint":"Press releases turn percentages into phrases. Ask the AI to convert them back, label each stat, and treat the survey as the only judge.","rubricLabels":["Labels stats 1, 2 and 3 Matches (1,204 is more than 1,200; 68% is about two out of three; 47% is nearly half)","Labels stat 4 Wrong with the survey's 22% (not one in three), and stat 5 Not in survey","Outputs only the 5 labeled lines using Matches, Wrong or Not in survey: no intro, no rewritten release, no PR advice"]}