Separate must-haves, preferences, and marketing claims
Author: AILesson6 min setupTested with:ChatGPTReviewed: 2026-08-28
Quick answer
Classify purchase statements by actual need and evidence instead of letting persuasive wording define value. Provide: Actual use and consequences, Features and marketing material, Decision and evidence rules. Expected result: A needs hierarchy, claim-evidence audit, and focused verification list.
1
Add your context
Your text stays in this browser. AILesson Prompts does not send it to a model or server.
2
Your prompt
Unfilled fields remain visible as placeholders, so you can still copy and edit the prompt
Audit the supplied product features and claims against the actual use.
Tasks, users, frequency, environment, important failures, alternatives, and budget:
[use]
Features, labels, slogans, specifications, reviews, prices, and source labels:
[claims]
Non-negotiables, preferences, prohibited assumptions, acceptable proof, and trial options:
[rules]
Classify each statement as must-have requirement, weighted preference, neutral descriptor, evidence-backed benefit, unverified marketing claim, irrelevant feature, or unclear. A feature is not a benefit until its connection to the stated job and consequence is shown. Decompose vague superlatives and proprietary labels into testable claims; do not assume certification, superiority, durability, health, safety, or compatibility from branding. Preserve contradictory evidence and source type. Return: needs hierarchy; claim-audit table with exact wording, category, relevance, evidence status, measurable proxy, and verification method; duplicated benefits; features that may create cost or complexity; questions for seller or manufacturer; and a short rule for ignoring distractions during comparison.
Private by defaultPrompt assembly happens locally in your browser. Avoid placing confidential information into any AI service unless your organization allows it.
From input to outcome
A worked example
See how concrete context turns this recipe into a usable result
Actual input
Actual use and consequences
Buy a blender for smoothies four mornings/week for two adults. Usually frozen berries, banana, yogurt, and seeds. Small kitchen, downstairs neighbor, hand-wash only. Current blender leaves berry chunks and is very loud. Budget USD 180. No need for hot soup, phone app, or single-serve travel cups.
Features and marketing material
Product page: ‘ProBlend Cyclone Intelligence,’ 1,400 peak watts, six Auto-IQ programs, aircraft-grade stainless blades, whisper-quiet enclosure, self-clean cycle, app recipes, 2.1 L jar, USD 169. Manual: do not run hot liquids; base 22 × 24 cm; jar and lid top-rack dishwasher safe. Retailer reviews: several call it powerful; three recent reviews mention high-pitched noise. No decibel result supplied.
Decision and evidence rules
Must fit a 25 × 25 cm counter area, crush the stated frozen ingredients, and stay within budget. Lower noise and easy hand cleaning are preferences. Marketing label alone is not proof. Can test texture, cleaning time, and noise in a 14-day return period. Do not infer blade material safety or durability from ‘aircraft-grade.’
Example output
Needs hierarchy: gates are footprint ≤ 25 × 25 cm, price ≤ USD 180, and demonstrated ability to blend the specified frozen mixture. Preferences are lower observed noise, short hand-cleaning time, and manageable capacity. Hot-liquid use, app recipes, and six programs are irrelevant here.
Claim audit: 22 × 24 cm footprint and USD 169 are specifications supporting two gates. “1,400 peak watts” is measurable but peak power alone does not prove smoothie texture; test one fixed recipe and inspect remaining chunks. “ProBlend Cyclone Intelligence” and “Auto-IQ” are proprietary labels until their actions and benefit are specified. “Aircraft-grade stainless” describes a claim about material but does not prove safety, sharpness retention, or life. “Whisper-quiet” is unverified and conflicts with several high-pitched-noise reports; compare with the current blender at the same position using the same phone meter only as a consistent proxy, not a laboratory value. “Self-clean” is relevant only if it reduces manual cleaning; time the full rinse and wipe process. Dishwasher-safe is not a benefit for this household.
Potential complexity: extra programs and app features add controls without serving the stated job; the 2.1 L jar may increase hand-cleaning burden. Seller questions: measured noise method, continuous rather than peak power, and whether frozen-ingredient use is explicitly covered by instructions and warranty. Distraction rule: ignore any feature that cannot be linked to smoothie texture, space, noise, cleaning, price, or supported life.
Why this works
1
Linking every feature to a real job prevents feature count from becoming a substitute for usefulness.
2
Testable proxies convert vague labels into questions that evidence can answer.
Check the result
Is every must-have tied to a concrete failure or consequence?
Are labels and superlatives separated from specifications, tests, and independent evidence?
Does each relevant uncertainty have a feasible verification method?
Use it with confidence
Frequently asked questions
Practical answers about when to use this recipe, what to provide, and where human review still matters
What should I prepare before using “Separate must-haves, preferences, and marketing claims”?
For “Separate must-haves, preferences, and marketing claims,” prepare Actual use and consequences, Features and marketing material, and Decision and evidence rules. Replace placeholders only with information you can verify. If a detail is unknown, preserve that uncertainty explicitly instead of asking the model to infer it.
When is the “Separate must-haves, preferences, and marketing claims” result not ready to use?
The result is not ready if it does not yet deliver the stated outcome—A needs hierarchy, claim-evidence audit, and focused verification list—from the supplied evidence, or if it relies on unresolved assumptions, missing approvals, or invented details. Use the checks as release gates: revise the source inputs or assign a named, authorized reviewer instead of polishing an unsupported output.
Which AI tools have recorded tests for “Separate must-haves, preferences, and marketing claims”?
The published test record for “Separate must-haves, preferences, and marketing claims” lists ChatGPT as of 2026-08-28. This confirms recorded runs, not guaranteed compatibility or identical results in later product versions. For another tool or version, keep every constraint visible and repeat the result checks before use.
ChatGPT