A competitor lists a $500 service. Another charges $1,500. Which one should shape your offer? Neither price means much without knowing what the buyer receives. This ChatGPT prompt for pricing research helps you gather sourced prices, compare scope and proof, identify buyer risk, and form hypotheses without copying competitors.
Use it with current public pricing pages or notes you have verified. The result is a decision-support document, not an AI-generated price recommendation. Research informs pricing. It does not determine pricing.
Why “What Should I Charge?” Fails
Imagine two operations consultants. One sells a diagnostic call. The other delivers a diagnostic, implementation plan, and month of support. Their public prices describe different commitments. Asking ChatGPT to average those prices erases the difference that matters.
The problem is not arithmetic. It is evidence quality. An AI assistant may also recall an outdated price, mistake a promotional rate for a standard fee, or confidently fill missing package details. None should become your pricing baseline.
What Pricing Research Can and Cannot Tell You
Use this method when developing a new service package, reviewing a subscription tier, or checking how comparable sellers present their offers. It can reveal observed ranges, typical inclusions, access levels, proof, and payment commitments.
Do not use competitor pages alone to infer willingness to pay, profit margins, conversion rates, or your optimal price. Those questions need internal financial data, buyer interviews, historical deals, or controlled tests. The U.S. Small Business Administration’s market research guidance distinguishes research on alternatives from direct research with customers.
Choose Truly Comparable Offers
Start with buyers, outcomes, delivery model, and geography. A self-serve course and a done-for-you consulting project may solve similar problems, but they are not direct substitutes. Neither are B2B retainers and consumer subscriptions.
- Direct comparable: similar buyer, outcome, service depth, duration, and market.
- Partial comparable: same core job, but material differences in scope or delivery.
- Weak comparable: overlaps on a feature, not the overall buyer commitment.
- Not comparable: different buyer, outcome, or economic model.
Collect enough genuine matches to notice patterns. Six to eight candidates may help a narrow consulting study, but no universal sample size validates a market. Keep partial comparables visible and analyze them separately.
The Price → Scope → Proof → Risk → Fit Framework
Price: Record the published amount, currency, billing unit, and whether it is standard or promotional. Scope: List deliverables, implementation, revisions, access, support, and duration.
Proof: Record relevant case studies, testimonials, client examples, qualifications, trials, or samples. A detailed, relevant case study may offer more useful evidence than twenty generic quotes. Rate proof Strong, Moderate, Limited, or Not visible, with reasons.
Risk: Note guarantees, refunds, cancellation terms, phased payments, and clear implementation boundaries. These features may reduce uncertainty; they do not prove higher conversion. Fit: Decide whether the offer belongs in your comparison at all.
Complete ChatGPT Prompt for Pricing Research
Copy the following master prompt and replace the bracketed fields. Supply public URLs, copied page text, pricing tables, screenshot transcriptions, or structured notes. If live search is unavailable, the analysis must use only your supplied evidence.
Ready-to-copy master prompt
ROLE: Act as a careful pricing-evidence researcher for a small business. Your job is to analyze comparable published offers, not set my prices.
BUSINESS CONTEXT: Business type: [type]. Own offer: [name and outcome]. Target buyer: [buyer]. Geography: [market]. Sales model: [B2B/B2C and sales channel]. Current price: [amount or Unknown]. Pricing question: [specific question]. Research objective: [range, packages, scope, proof, risk, or hypothesis].
MY OFFER: Candidate prices: [amounts]. Deliverables: [details]. Duration: [duration]. Meetings/access: [details]. Implementation: [included/not included]. Customization: [level]. Revisions and support: [details]. Guarantee: [terms or Unknown]. Proof: [evidence]. Payment terms: [terms]. Unknown fields must stay Unknown.
COMPARABLE INPUTS: Paste each offer in this format, repeating as needed: Provider [name]; source URL [link]; date checked [date]; offer [name]; price [quoted price or Unknown]; currency [currency]; unit [per project/month/year/user/hour/etc.]; commitment [minimum term]; taxes [included/excluded/Unknown]; promotion [standard/temporary/Unknown]; setup and overage fees [details]; deliverables [details]; duration [details]; meetings/support/access [details]; customization [details]; proof [source details]; guarantee, refund, trial, or cancellation terms [details]; payment terms [details]; positioning [details]; availability [current/archived/Unknown]. You may also use labeled copied pages or structured notes.
SOURCE RULES: Work source first: Source → Extract → Normalize → Compare → Interpret. Prefer official pricing pages, official offer pages, public official proposals, official marketplace listings, then reputable sources directly quoting provider terms. Avoid affiliate roundups, forum comments, and broad average-price articles as primary price evidence. If browsing is available, verify public pages now and record the check date; otherwise do not imply a live check. If a URL cannot be opened, state that and request pasted evidence. Do not invent a citation, source excerpt, or check date.
PRICING RULES: Every material price requires a source, check date, currency, billing unit, offer name, tax status if known, and promotion status if known. Label prices Observed, Calculated, or Estimated. Show conversion math, original units, and payment commitment; a $1,188 annual fee can be shown as $99 monthly equivalent, billed annually. Do not silently mix one-time, hourly, retainer, subscription, per-user, or usage prices. Never mix currencies without a dated, verified exchange rate. Keep unverified amounts outside observed price ranges. For missing prices write Price not verified.
COMPARISON RULES: Classify each offer Direct Comparable, Partial Comparable, Weak Comparable, or Not Comparable using buyer, outcome, scope, duration, service depth, delivery, and geography. Compare Price → Scope → Proof → Risk → Fit. Check revisions, implementation, support, speed, reporting, setup fees, minimum commitment, annual prepayment, usage limits, add-ons, cancellation rules, and hidden conditions where visible. Label proof Strong, Moderate, Limited, or Not visible and justify it. Note positioning only when it changes comparability. Flag outliers; investigate scope, buyer, promotion, proof, or delivery differences before interpreting them. Never average unrelated offers. Use mean or median only if units, comparability, and sample justify it.
NO INVENTION: Never invent competitor prices, discounts, margins, conversion rates, revenue, sales volume, market share, customer demographics, demand elasticity, willingness to pay, internal costs, hidden scope, motivations, or reasons for pricing choices. If unsupported, write Not supported by available pricing evidence. Do not treat advertised prices as realized transaction prices or as proof of willingness to pay. Do not recommend my price from a competitor midpoint.
REQUIRED OUTPUT, IN ORDER:
- Research Quality Check: counts of offers by comparability, missing prices, freshness, and gaps.
- Comparable Pricing Table: Provider | Offer | Price | Pricing Model | Scope | Proof | Risk Reduction | Comparability | Source | Date Checked. Show currency, unit, promotions, and tax uncertainty in relevant cells.
- Observed Price Range: direct and partial comparables separately; call it an observed competitor range, not validated willingness to pay.
- Scope Differences: explain material differences and price-model mismatches.
- Proof Differences: classify strength and relevance with evidence.
- Buyer-Risk Differences: identify guarantees, payment friction, and support.
- Packaging Patterns: describe evidence-backed tiers and access levels.
- Outliers: show unusual prices and possible structural explanations.
- Evidence Gaps: list unknown facts and unsupported conclusions.
- Pricing Hypotheses: maximum three; each needs evidence, rationale, uncertainty, and a validation method.
- Next Validation Steps: maximum five prioritized actions, each with its decision purpose.
INTERPRETATION: Mark every important conclusion Observed, Inferred, or Test. Use a final decision label: Enough Evidence to Form Hypothesis; Need More Comparable Offers; Need Buyer Research; Need Financial Modeling; Ready for Controlled Test; or Not Comparable. State which evidence would change your conclusion. Do not output a Recommended Price.
SAFETY AND FOLLOW-UP: Use only public or legitimately accessible information. Do not request confidential proposals, customer contracts, personal data, or bypassed access restrictions. If evidence is missing, produce the partial comparison, flag the gaps, and ask only for the most consequential missing sources. Finish with one specific next validation action, not a definitive pricing decision.
How to Personalize the Prompt
Keep offer descriptions consistent across rows. A blank field is not permission to guess. Write Unknown when the official source omits a detail.
| Prompt Field | What to Enter | Example |
|---|---|---|
| Business Type | Business and delivery model | Independent operations consultant |
| Target Buyer | Who signs and pays | U.S. small service firms |
| Geography | Country or service market | United States, remote |
| Own Offer | Buyer outcome | Process diagnostic |
| Current Price | Existing amount or Unknown | Unknown |
| Candidate Prices | Internal options under review | $500 / $900 / $1,500 |
| Offer Scope | Length, access, implementation | 90 minutes plus action plan |
| Comparable URLs | Public offer pages | Six to eight relevant URLs |
| Price | Advertised amount and currency | $750, publicly listed |
| Pricing Model | Billing unit and commitment | Per project, one-time |
| Deliverables | Actual included work | Audit and follow-up call |
| Proof | Relevant public evidence | Two case studies |
| Risk Reduction | Terms that reduce uncertainty | Defined revision policy |
| Date Checked | Day you verified each page | Date of your source review |
| Research Goal | Question to investigate | Compare diagnostic package scope |
How to Verify Prices and Normalize Scope
Open each public source yourself. Save the exact offer name, checked date, currency, taxes if stated, billing period, promotion status, and purchase commitment. Official pricing and service pages outrank third-party summaries. OpenAI’s guidance on reviewing search sources recommends checking whether citations actually support the claims.
A $99 monthly equivalent from a $1,188 annual payment is not a cancellable $99 monthly subscription. Likewise, a $500 diagnostic cannot be directly averaged with a $1,500 implementation package. Normalize access, deliverables, revisions, speed, duration, and aftercare first.
Separate observed published prices from transparent calculations. Keep estimated and stale prices outside the core range. Mark “tax treatment not verified” when the listing is silent. If currencies differ, retain originals; any conversion needs a verified rate and date.
Filled Example: $500 vs $900 vs $1,500 Consultant Offers
Illustrative example only: the following competitor observations are invented teaching data, not real firms or verified market prices. A consultant has four years of experience, several testimonials, limited quantified case studies, and eight available consulting hours per week. The target buyer is a small U.S. service business.
| Candidate Package | Deliverables | Buyer Commitment | Open Question |
|---|---|---|---|
| $500 | 90-minute call and short written summary | One-time; no follow-up | Is diagnosis alone clearly valuable? |
| $900 | Diagnostic, prioritized action plan, follow-up call | One-time project | Do buyers value prioritization and access? |
| $1,500 | Diagnostic, action plan, 30-day implementation support | One-time; support window | Can scope and capacity be protected? |
Customized prompt excerpt: “Compare six to eight U.S. operations consulting offers for small service businesses. Keep diagnosis-only projects separate from offers with implementation support. Evaluate my $500, $900, and $1,500 candidate scopes. Do not infer a recommended price.”
Sample AI output, illustrative: Research Quality Check: seven fictional offers inspected; three direct, three partial, one weak comparable. One hides its price. All listed amounts are hypothetical, so none are verified market observations.
Observed within the fictional dataset: diagnosis-only comparables tend to appear at the lower end; offers with implementation or access appear higher. Some premium packages show detailed case studies or clearer support boundaries. These are fictional dataset patterns, not claims about the actual consulting market.
Inferred: the $1,500 package may need stronger outcome proof and a defined support limit. The $900 package could be distinguished through its prioritized plan, rather than presented as a midpoint. Test: interview five qualified prospects or compare package reactions in controlled sales conversations, then review the cost and capacity of delivering each option.
Different business, same boundary: a solo ecommerce seller considers $29, $49, and $79 monthly subscription tiers. Competitor tier prices may reveal included features and usage limits. They cannot establish the seller’s gross margin, retention, demand, or willingness to pay.
How to Read the Output Without Copying Competitors
An observed range is a description of published offers. It is not a clearing price or a buyer’s maximum budget. Label it Observed competitor range, not validated willingness to pay. Do not let a neat spreadsheet turn weak comparables into false certainty.
Next, separate the handoffs. Pricing research gathers evidence. Pricing strategy decides how that evidence affects your offer. Financial planning checks margin, cash flow, and delivery capacity. Offer testing measures what happens when a hypothesis reaches real buyers.
For the business decision after evidence collection, turn your pricing research into a pricing strategy without treating competitor numbers as instructions.
If the question extends beyond price, conduct broader competitor research into positioning, channels, and market signals.
Before implementing a material change, model the financial impact of a price change using your own cost and capacity figures.
Once you have a defensible hypothesis, test a pricing hypothesis with controlled offer experiments rather than declaring research to be validation.
What to Do When Competitors Hide Their Prices
“Contact us” and “Custom quote” are pricing disclosures about the sales process, not price amounts. Record Price not verified. You can still compare target buyers, service scope, proof, consultation requirements, contract length, and purchase friction.
Do not reverse-engineer a secret price from a competitor’s client list or testimonials. If your direct-comparable sample has too many hidden prices, label the study Need More Comparable Offers and continue with offer-structure analysis.
Common Pricing Research Mistakes
- Asking “What should I charge?” invites an unsupported decision. Ask for sourced comparisons and hypotheses instead.
- Trusting AI memory or old pages. Prices change. Open official sources and log each check date.
- Mixing billing units, currencies, or promotions. A discounted annual commitment is not an ordinary monthly fee. Label every difference.
- Averaging unrelated offers. Different buyers, delivery depth, or geography distort the midpoint. Classify fit first.
- Ignoring proof and buyer risk. A guarantee, relevant case study, or extra access may explain meaningful scope differences. Record each.
- Assuming expensive means premium. Price alone does not prove better positioning or outcomes. Look for actual supporting evidence.
- Inventing competitor economics or buyer motives. Public offer pages cannot reveal margins, conversions, or pricing intent. Mark unsupported claims.
- Confusing desk research with validation. Public prices do not reveal willingness to pay. Test against customer evidence and your own capacity.
Output Review Checklist
- Does every reported price have an accessible source and actual check date?
- Are currency, pricing unit, tax uncertainty, and promotion status explicit?
- Are direct, partial, weak, and non-comparable offers separated?
- Does the analysis account for implementation, access, proof, guarantees, and cancellation risk?
- Are Observed, Inferred, and Test conclusions clearly separated?
- Are outliers explained rather than removed to make a prettier average?
- Are missing facts labeled instead of guessed?
- Does the report end with no more than three hypotheses and one actionable validation step?
7-Day Pricing Research Plan
| Day | Task and Time | Expected Output |
|---|---|---|
| 1 | Define your pricing question — 20 minutes | One decision question |
| 2 | Set buyer, outcome, delivery, geography filters — 20–30 minutes | Comparability criteria |
| 3 | Find candidate offer pages — 45–60 minutes | Up to 5–10 sources if relevant |
| 4 | Verify terms, dates, scope, and units — 45 minutes | Clean evidence table |
| 5 | Run the master prompt — 30–45 minutes | Source-backed analysis |
| 6 | Review up to three hypotheses — 30 minutes | Qualified hypotheses and gaps |
| 7 | Choose a buyer interview, sales test, or financial model — 30 minutes | One validation action |
Frequently Asked Questions
Can ChatGPT research competitor pricing?
Yes, when it has current, accessible sources or pages you provide. Verify links and dates yourself. If search is unavailable, it should analyze supplied notes without claiming the prices are live.
Can ChatGPT tell me what to charge?
Not from competitor prices alone. A $900 consulting package might include a follow-up call while another $900 offer includes implementation. Your costs, evidence, capacity, and buyer response still matter.
How many competitor offers should I compare?
Use enough close matches to see recurring patterns. Four highly relevant sources can be more useful than twelve unrelated listings. A small sample supports hypotheses, not a market-wide claim.
What if competitors show no prices?
Record Price not verified. Compare public scope and risk reducers instead. For example, a custom-quote agency might still disclose onboarding, revisions, contract duration, and case studies.
Should I average monthly and annual prices?
No. First normalize the units and retain the commitment. A $1,188 annual subscription equals $99 per month mathematically, but it remains billed annually. Only calculate summary statistics for genuinely comparable offers.
Do competitor prices show willingness to pay?
No. They show what a seller advertises, not what comparable buyers accepted. Use sales conversations, transactions, interviews, or controlled tests to examine willingness to pay.
Conclusion
A useful ChatGPT prompt for pricing research turns scattered public prices into evidence you can inspect. Gather comparable offers, verify the terms, and compare scope, proof, risk, and fit. Your output should be a short list of testable hypotheses, never a magical recommended price.
Start with one clearly defined offer, collect its closest public comparables, and run the master prompt. Then choose one real-world validation step.
