Find the real patterns in employee reviews, with review numbers

About this exercise

You're deciding whether to apply to Orchard Lane Credit Union and pasted 8 employee reviews. You want exactly 3 numbered themes in plain text, each mentioned in at least 2 reviews, each under 12 words, and each ending with the review numbers in parentheses, like "(2, 5)". Skip one-off complaints and add nothing else.

Common mistakes when you ask AI for this

• Not asking for review numbers. The AI summarizes 'what employees say' with no way to check which reviews said it. Try: Add: "End each theme with the review numbers, like (2, 5)."
• Treating a rant as a pattern. Colorful one-off comments stick out, and the AI may promote them to a theme. Try: Say: "Only themes mentioned in at least 2 reviews."
• Getting a verdict. The AI ends with whether you should apply, which you didn't ask for. Try: Say: "Output only the 3 themes."
• Not saying how many. The AI decides the length itself and usually returns more themes than you can use. Try: Be exact: "exactly 3 numbered themes."
• Getting markdown back. AI defaults to bold labels and headings, which show up as stray symbols in your job search tracker. Try: Ask for "plain text, no bold, no headings."

What a strong prompt includes

• Says exactly how many
• Asks for review numbers
• Requires repeat mentions
• Limits each line
• Asks for plain text

Practice this now ↑

Related exercises

• Accept a job offer and confirm the exact terms (Writing AI · Job seekers)
• Alert sales to a competitor's price change, facts only (Research AI · Product managers)
• Answer a 'where is my order' email with the real new date (Writing AI · Customer support teams)
• Answer a big discount ask with two priced options (Writing AI · Sales and customer success)
• Answer a buyer's HOA questions from the bylaws (Research AI · Real estate agents)