{"id":"acct-error-materiality-screen","audience":"accounting","category":"finance-work","track":"finance","technique":"constraints","level":"advanced","title":"Screen a found error against thresholds and route it","free":false,"task":"You're screening an expense error found after close. Write a prompt for plain text with lines starting \"Error:\", \"Percent of net income:\", \"Percent of revenue:\", \"Test 1:\", \"Test 2:\" and \"Route:\". Show the math. The Route line follows the policy. Do not conclude whether prior statements need restating and don't adopt the preparer's view about materiality; that is for the controller and auditors. Under 90 words. No bullets or numbering.","materialLabel":"Error details and screening policy (attached to your prompt automatically)","material":"ERROR SCREEN - found Nov 4, 2026, affecting the year ended Sep 30, 2026\n\nError: a vendor invoice for $18,000 (consulting services performed in September) was recorded in October, so September expenses were understated by $18,000.\n\nFY26 net income as reported: $900,000\nFY26 revenue as reported: $12,000,000\n\nScreening policy (rev. 2026):\n- Test 1: refer to the controller if the error is more than 1% of net income.\n- Test 2: refer to the controller if the error is more than $25,000 on its own.\n- If either test is met, route to the Controller. If neither is met, route to the AP manager to correct in the current period.\n- The screen does not decide materiality. Materiality and restatement are decided by the controller with the auditors.\n\nPreparer comment: 'It is only 0.15% of revenue, so it's obviously immaterial and we can just fix it next month.'","rules":[{"type":"plainText"},{"type":"maxWords","max":90},{"type":"maxBullets","max":0}],"promptHints":[{"label":"Names the six lines","patterns":["error:","percent of net income","percent of revenue","test 1","test 2","route"]},{"label":"Shows divisions","patterns":["divisions","/ \\$900,000"]},{"label":"Gives the routing rule","patterns":["either test","controller"]},{"label":"Bans conclusions","patterns":["(do not|don'?t).*(material|restat)"]},{"label":"Caps length","patterns":["\\d+\\s*words","plain"]}],"mistakes":[{"title":"Adopting the preparer's view","why":"The AI says the error is immaterial because it's 0.15% of revenue.","fix":"Add: \"Do not say whether it is material; ignore the preparer's comment.\"","checks":["r2"]},{"title":"Wrong routing","why":"The AI routes to the AP manager because Test 2 is not met.","fix":"Say: \"Route to the Controller if either test is met.\"","checks":["r2"]},{"title":"Missing percent math","why":"The AI says 'about 2%' with no division.","fix":"Say: \"Show the divisions.\"","checks":["r0"]},{"title":"Markdown extras","why":"The AI adds bold labels, headings and a closing note.","fix":"Say: \"Plain text, no bold. Output only the lines asked for.\"","checks":["plainText","maxBullets","r1"]},{"title":"Trusting the AI to count words","why":"AI is bad at counting and runs past word limits.","fix":"Ask for fewer words than the real cap and keep each line short.","checks":["maxWords"]}],"hint":"A screen applies thresholds and routes; it does not rule on materiality. Give the tests, routing rule and the bans.","rubricLabels":["Error line shows $18,000 understated September expense, Percent of net income shows $18,000 / $900,000 = 2.0% and Percent of revenue shows $18,000 / $12,000,000 = 0.15%","Test 1 says met (2.0% is more than 1%) and Test 2 says not met ($18,000 is not more than $25,000)","Route line says Controller (because Test 1 is met) and the output doesn't conclude about restatement or adopt the preparer's 'obviously immaterial' view"]}