01 / START HERE
A useful way to approach it.
Before estimating a market, define the customer's problem and location. 'Small businesses' is too broad for a useful demand question. 'Independent tutors serving one city who need evening classroom space' gives you a population, a need, and a geography you can investigate. AI can help organize that investigation without turning an estimate into a fact.
This is secondary research using public material or sources you provide. It does not conduct interviews, purchase proprietary datasets, or prove willingness to pay. When current retrieval is needed, use enabled web search in signed-in o1 and inspect the cited material. The fictional example illustrates an estimate structure; its quantities are authored assumptions, not measured demand.
02 / THE PROCESS
Go from question to next step.
- 01
Bound the market question
Choose one customer segment, one use case, and one location or distribution area. State whether you need to compare segments, understand an existing problem, or estimate a possible opportunity. Define your units: people, establishments, subscriptions, or annual spending. Mixing those units makes even careful arithmetic difficult to interpret.
- 02
Plan the evidence search
List the information that would support the decision: population counts, existing alternatives, reported buying behavior, access constraints, and relevant trends. Match each question to appropriate public sources. Record publication date, geography, and method where available. If evidence concerns a broader population, describe how it differs from your intended customers.
- 03
Create an assumption ledger
Put sourced quantities in one section and your assumptions in another. An establishment count is not automatically a count of likely buyers. For every estimated adoption rate, purchase frequency, or price, explain why it is provisional. Use a range only when you can state where its endpoints came from.
- 04
Compare plausible scenarios
Calculate a small set of scenarios with the same formula and units. Show how changing one uncertain input affects the result. Avoid a single impressive market number that hides your assumptions. Ask the assistant to check the arithmetic, then verify it independently before using the memo in a pitch or budget discussion.
- 05
Choose the next validation step
Identify the assumption most likely to invalidate the idea. Design a modest next step such as interviews you arrange, an offer description to review with prospective customers, or a test of access constraints. The research memo should tell you what to learn next. Public attention, search interest, and a large population are not proof of buying intent.
03 / YOUR STARTING POINT
A prompt you can make your own.
Replace the bracketed fields. Add only information you have permission to share. This is a reusable text prompt; it does not run a task from this page.
Create a secondary-research plan for {{customer}} with {{problem}} in {{geography}} during {{period}}. Use {{sources}} and current primary sources only if enabled web search retrieves them. Separate sourced facts, assumptions, and unknowns. Define the counting unit and show any estimate formula. Compare low, middle, and high illustrative scenarios without inventing measurements. Explain source coverage and date limitations. End with the most fragile assumption and three questions for validation that I would arrange myself.04 / SEE THE SHAPE OF IT
Illustrative classroom-space opportunity memo
Fictional scenario with authored quantities for arithmetic practice; no market size or demand has been measured.
| Input / scenario | Authored assumption | What to verify |
|---|---|---|
| Potential customers | 100 fictional local tutors | Find an appropriate dated count |
| Initial interest | 10–20 tutors | Interest is not a purchase commitment |
| Use frequency | 2 evenings each month | Ask about actual schedules |
| Scenario volume | 20–40 tutor-evenings per month | Check availability and repeat use |
The multiplication is inspectable, while the important inputs remain clearly provisional. The next job is learning whether those assumptions describe real buyers.
05 / KEEP YOUR JUDGMENT
What to watch before you use it.
Using a broad industry total as demand
A large category can include buyers you cannot serve. Narrow the segment and explain the remaining gap before estimating your opportunity.
Combining incompatible reports
Different geography, dates, and definitions can make totals incomparable. Retain those differences rather than averaging them into false precision.
Letting the answer invent missing data
Ask for an evidence gap when a quantity is unavailable. A plausible figure without provenance does not become a reliable input.
06 / THE FINAL PASS
A little review goes a long way.
Your checks stay on this page. They do not approve or run a task.
07 / GOOD QUESTIONS
Before you begin.
Can this replace customer interviews?
It can prepare better questions, but it does not collect primary evidence from prospective customers. Use public research to understand the context, then arrange your own validation when the decision depends on actual preferences or willingness to pay.
What if two market estimates disagree?
Compare their definitions, dates, geography, and methods before comparing totals. They may describe different markets. If you cannot reconcile them, show both with their limitations and avoid presenting a blended number as a measured fact.
Should I start with market size or the customer problem?
Start with the decision and the customer problem. Define the segment before counting it. A smaller, clearly described group with a testable need is more actionable than a broad total that does not explain who might buy.