Analyze this business decision as a rigorous operating memo. Do not invent market data, customer evidence, financial results, competitor facts, or stakeholder views. Label missing evidence and distinguish facts from assumptions. Decision to make: [DECISION] Business context: [BUSINESS CONTEXT] Available evidence: [EVIDENCE] Options already under consideration: [OPTIONS] Constraints and decision criteria: [CONSTRAINTS AND CRITERIA] Financial or operating baseline: [BASELINE METRICS] Work in this order: 1. Restate the decision in one sentence, identify the decision owner, and specify the reversible versus irreversible elements. Do not expand the scope into a broad company strategy. 2. Build a fact base from my evidence. List the important unknowns and say how each could change the recommendation. 3. Evaluate each option against the stated criteria, including expected upside, cost, time to impact, execution complexity, customer impact, dependencies, downside risk, and leading indicators. Use ranges only when I supplied a basis; otherwise describe the uncertainty qualitatively. 4. Identify the likely counterargument to your recommendation and the conditions under which a different option would win. 5. Recommend one option, or a staged experiment if evidence is insufficient. Make the recommendation conditional where necessary rather than falsely certain. Output exactly: an executive recommendation of 150 words or fewer; a decision table; an assumptions and evidence-gap register; a 30-day validation plan with owner, action, metric, threshold, and decision date; and a short list of risks with mitigations. Before answering, check that the recommendation follows from the table and that no numerical claim lacks a source in my inputs. Ask up to 3 clarifying questions only if a required input is missing.
Fill in
| Placeholder | What to enter | Example |
|---|---|---|
| [DECISION] | State the specific decision, decision deadline, and person or group who will decide. | Choose whether to add self-serve annual billing by October 15; VP Product and CFO decide. |
| [BUSINESS CONTEXT] | Describe the product, customer, business model, market stage, and why the decision matters now. | B2B workflow SaaS for 80-person operations teams; $3.2M ARR; monthly contracts today; sales-led mid-market motion. |
| [EVIDENCE] | Paste relevant research, customer feedback, experiment results, notes, or source links with key facts. | 14 of 42 closed-won customers asked about annual billing; three sales calls cite procurement preference; no tested pricing data. |
| [OPTIONS] | List the realistic choices, including doing nothing when it is a genuine option. | A: annual plan at 15% discount; B: annual plan at 10% discount with implementation fee; C: keep monthly plans only. |
| [CONSTRAINTS AND CRITERIA] | State budget, timing, team capacity, risk limits, and the criteria that should govern the choice. | Two engineers for six weeks; cash collection and retention matter most; no discount above 15%; legal review required. |
| [BASELINE METRICS] | Provide current metrics such as revenue, conversion, retention, cost, capacity, or cycle time with dates. | Monthly logo churn 1.8%; average contract $7,900 ARR; sales cycle 52 days; implementation capacity 12 customers per month. |
How to use
- State one decision rather than asking for a general business assessment.
- Paste the evidence as notes, links, numbers, and dated findings; mark opinions as opinions.
- Inspect the assumptions register and remove any option the team cannot actually execute.
- Follow up with: “Turn the 30-day validation plan into a one-page experiment brief for the product and sales owners.”
Variations
Market entry
Use this when choosing a new segment or geography.
Assess whether to enter [TARGET MARKET] with [PRODUCT]. Use [CURRENT CUSTOMER EVIDENCE], [MARKET DATA], [COMPETITOR NOTES], and [RESOURCE LIMITS]. Define the entry thesis, ideal customer profile, buying trigger, route to market, and reasons the segment might reject us. Compare launch, pilot, and no-entry options in a table. Output a recommendation, evidence gaps, a 90-day pilot design, and explicit stop criteria. Do not claim market size or competitor capability without supplied evidence.
Pricing decision
Use this when changing packaging, price, or discount policy.
Analyze a proposed pricing decision using [CURRENT PRICING], [PROPOSED CHANGE], [CUSTOMER RESEARCH], [WIN-LOSS DATA], and [FINANCIAL BASELINE]. Separate willingness-to-pay evidence from internal revenue targets. Evaluate conversion, expansion, churn, sales friction, billing complexity, and grandfathering risk. Output a pricing hypothesis, a segmented impact table, an experiment plan, customer communications risks, and recommendation. Flag where causal impact cannot be inferred from the supplied data.
Root cause review
Use this when a key metric has declined.
Investigate the decline in [METRIC] from [STARTING LEVEL] to [CURRENT LEVEL] over [TIME PERIOD] using [DATA AND EVENTS]. Create a timeline, distinguish correlation from likely causal mechanisms, and rank hypotheses by evidence strength and business impact. Output a root-cause tree, the smallest tests that could disprove each leading hypothesis, and immediate containment actions. Do not call a cause confirmed unless the evidence supports it.
Tips
- Frame the decision so a person can say yes or no; “improve growth” is a goal, not a decision.
- Date every metric and research finding, because stale evidence can make an option look better than it is.
- Include doing nothing when it preserves capacity or avoids an irreversible commitment.
- Define a threshold before running an experiment so the result cannot be reinterpreted after the fact.
FAQ
Can AI perform a market analysis without research?
It can structure a hypothesis and identify research gaps, but it should not make market-size or competitor claims without credible supplied sources.
How many options should I include?
Usually two to four actionable alternatives are enough. More options often conceal that the decision criteria have not been set.
What makes a recommendation useful?
It names the decision, the evidence supporting it, the key risk, and the condition that would change the answer.