Rescore a feature backlog with RICE after a new estimate

About this exercise

You're picking next quarter's first build using RICE (reach x impact x confidence ÷ effort). Write exactly 4 numbered lines, one per feature, each showing the math and score, then one unnumbered last line: Build first: [feature]. In this morning's standup, engineering re-estimated Slack alerts at 1.5 person-weeks (the sheet isn't updated yet). Ignore opinions that aren't in the scoring, plain text, under 90 words.

Common mistakes when you ask AI for this

• Using the old estimate. The AI scores Slack alerts on 3 weeks and picks Dark mode. Try: Add: "Slack alerts effort is now 1.5 weeks, per today's standup."
• Not asking to see the work. Without the math you can't spot a slipped decimal. Try: Add a line to copy: "1. Bulk export: 2,000 x 2 x 0.8 ÷ 4 = 800"
• Letting opinions in. The AI weighs the CEO's ask or design's hunch into the pick. Try: Add: "Pick on score only. Ignore opinions not in the inputs."
• Trusting the AI to count words. AI is bad at counting. Ask for a total word limit and it often lands over it. Try: Cap each line: "each line under 16 words."
• Forgetting where it goes. AI loves markdown tables and bold, which show up as clutter in a planning doc comment. Try: Add: "Plain text, no bold, headings or tables."

What a strong prompt includes

• Uses the new estimate
• Defines the score
• Asks to show the math
• Asks for the pick last
• Asks for plain text

Practice this now ↑

Related exercises

• Alert sales to a competitor's price change, facts only (Research AI · Product managers)
• Approve or return expense claims against the policy (Productivity AI · Finance and accounting)
• Approve part of an overtime request with the math (Writing AI · Managers and team leads)
• Average daily call activity after fixing a dialer glitch (Data AI · Sales and customer success)
• Average handle times after removing outage calls (Data AI · Customer support teams)