Verified against Perplexity Pro · 2026-08-05
Compare two specific purchase options against your own stated priorities, not generic pros and cons
A buying-decision prompt that requires the comparison to be weighted against a stated, ranked list of what actually matters to the buyer, rather than a generic feature-by-feature rundown that treats every difference as equally important.
The prompt
Ready to copy — highlighted parts are example details you can swap.
Help me decide between these two specific options. I don't want a generic pros-and-cons list — I want the comparison weighted against what actually matters to me, in the order I've ranked it. OPTIONS TO COMPARE The Framework Laptop 16 (AMD Ryzen 7040 configuration) vs The Lenovo ThinkPad P16s Gen 3 MY PRIORITIES, RANKED (most important first) 1) repairability and upgradeability, 2) battery life for a full workday, 3) build quality, 4) price MY CONSTRAINTS Must be available with at least 32GB RAM and ship to Canada WHAT I'M NOT WORRIED ABOUT Gaming performance — I don't game on this machine PROCESS 1. For each of my ranked priorities, state what each option actually offers, with a source for any factual claim (a spec, a price, a review-based reliability claim) — not a vague "generally considered good for X." 2. Check both options against my hard constraints first. If either option fails a hard constraint outright, say so plainly and explain that the remaining comparison is now academic for that option unless the constraint can be worked around, and say how. 3. Do not spend meaningful space comparing the two on Gaming performance — I don't game on this machine — a short acknowledgment that they're comparable there is enough; don't let a difference I said doesn't matter drive the recommendation. 4. If the two options are genuinely close on my top priority, say so explicitly rather than manufacturing a tiebreaker — being told "these are close enough that it's a coin flip on your top priority" is more useful than a confident recommendation built on a difference too small to matter in practice. RECOMMENDATION End with a direct recommendation given everything above, and — separately — name the one thing that would flip that recommendation if it turned out to be wrong (a price change, a spec I misremembered, a priority I ranked differently than I actually feel about it).
Customize
Optional — swap in your own details for the highlighted parts above.
Why this works
A generic pros-and-cons list treats every point of difference between two options as roughly equal in importance, which produces a comparison that reads as thorough but doesn't actually help a decision, because the reader still has to mentally re-weight every bullet point against what they personally care about — work the prompt should be doing, not offloading back onto the person who asked for help deciding. Requiring the comparison to walk through the buyer's own ranked priorities in order, rather than a neutral feature list, forces the structure of the output to match the actual structure of the decision, so the most important factor gets the most attention instead of getting one bullet point buried in the middle of a longer list alongside far less important ones. Checking hard constraints first and explicitly declaring the rest of the comparison "academic" for a failing option prevents a specific and common failure of comparison prompts: continuing to earnestly compare battery life and build quality between two options when one of them doesn't even ship with enough RAM to meet a stated requirement, which wastes the reader's attention on a decision that was already made by the constraint. And instructing the model to admit when two options are genuinely close on the top priority, rather than manufacturing a tiebreaker, matters because a confident-sounding recommendation built on a difference too small to be real in practice is actively worse than an honest 'these are close' — it gives the reader false certainty about a choice that, if the truth were stated, they might reasonably decide either way on grounds the comparison can't see, like personal aesthetic preference.
Verified against
Perplexity Pro Sonar Pro · 2026-08-05
Changelog
- 2026-08-05 — Initial publish, verified against Perplexity Pro Sonar Pro search.
Need this built into your business?
If a prompt isn't enough — what Scult builds, built and maintained for you — that's Scult's day job.
EXPLORE WHAT SCULT BUILDS
