Make a decision brief
Compare a small set of personal options without false precision.
Complete prompt
Task
Compare a small set of personal options without false precision.
Instructions
Help me decide between [OPTIONS] for [GOAL]. Use these criteria and weights: [CRITERIA]. Ask only for missing information that could change the ranking. Show evidence, assumptions, downside, and a reversible next step. Do not present a score as an objective truth.
Expected output
A compact comparison table followed by a conditional recommendation.
Checks
- Weights are visible.
- Evidence and inference are separated.
- The next step is reversible.Expected output and checks
Expected output
A compact comparison table followed by a conditional recommendation.
Checks
- Weights are visible.
- Evidence and inference are separated.
- The next step is reversible.
Worked example
Make a decision brief: recorded example
Original input
Task
Compare a small set of personal options without false precision.
Instructions
Help me decide between Fictional halls A:$80,capacity30,transit8/10,accessibility10/10; B:$100,capacity40,transit10/10,accessibility unknown; C:$60,capacity20,transit6/10,accessibility6/10. for Choose a hall for24 workshop participants, maximum budget$110. Capacity24 is mandatory.. Use these criteria and weights: Weights: cost40%,transit30%,accessibility30%. Cost scores A8/10,B6/10,C10/10. Use these supplied scores, not a new scoring formula. Accessibility is a mandatory prebooking check.. Ask only for missing information that could change the ranking. Show evidence, assumptions, downside, and a reversible next step. Do not present a score as an objective truth.
Expected output
A compact comparison table followed by a conditional recommendation.
Checks
- Weights are visible.
- Evidence and inference are separated.
- The next step is reversible.
What the checks found
- Recalculation: A scores 8.6; B ranges from 5.4 to 8.4; C fails the 24-person capacity minimum.
- The opening sentence says B could overtake A, contradicting the correct table and later paragraph.
Revisions after the first output
- Editor corrected the opening sentence and removed its obsolete self-correction. The first response remains available.
Original output
Hall A is the provisional choice; Hall B could rank higher if its accessibility score is above 7⅓/10, but neither should be booked until accessibility is checked.
Weights: cost 40%, transit 30%, accessibility 30%. These express your preferences, not an objective measure of hall quality. Scores below use only the supplied scoring scale.
| Hall | Supplied evidence | Mandatory capacity and budget checks | Preference-weighted score /10 | Downside or uncertainty |
|---|---|---|---|---|
| A | $80; capacity 30; cost 8, transit 8, accessibility 10 | Pass: at least 24 places and no more than $110 | 0.4×8 + 0.3×8 + 0.3×10 = 8.6 | $20 more than C, though C is ineligible; transit below B. Supplied accessibility score still requires a prebooking check. |
| B | $100; capacity 40; cost 6, transit 10; accessibility unknown | Pass | 5.4 + 0.3×accessibility score; between 5.4 and 8.4 if the missing score is on the same 0–10 scale | Accessibility unknown; costs $20 more than A. |
| C | $60; capacity 20; cost 10, transit 6, accessibility 6 | Fail: capacity below 24; budget passes | Not ranked because it fails the capacity requirement | Four too few places; low price cannot compensate for mandatory capacity. |
Conditional recommendation: choose A if its accessibility check meets the workshop’s needs. Based on the supplied scores, A’s 8.6 exceeds B’s maximum possible 8.4; the opening threshold claim is therefore incorrect. B cannot outrank A on this scoring scheme, but could become the viable choice if A fails the mandatory accessibility check and B passes. If both fail, neither is suitable.
Evidence versus inference: prices, capacities, and scores are fictional supplied facts. The weighted totals and A’s provisional preference are calculations and judgments under your chosen weights. A 0.2-point advantage over B’s best case should not be treated as a precise real-world quality difference.
Missing information that could change the viable ranking: what accessibility features do participants require, and do A and B meet those requirements on verification?
Reversible next step: request accessibility details from both halls and keep A as the provisional selection, without paying or committing to a booking.