# Do-it-yourself playbook: challenge a founder decision without outsourcing it

**Scope:** US pre-seed to Series A businesses, usually 2–50 employees. Evidence reviewed 28 October 2026. Advice is about the *process*, not a promise of an outcome.

## Objective and working assumptions

Use this programme when a founder must choose whether to pivot, accept a financing offer, hire a senior leader, enter a market, or reduce spend. The objective is a *defensible decision process* within the runway and confidentiality constraints—not a guaranteed correct answer. Keep a decision owner, an explicit stop date and one shared source of truth.

**Indicative project timings below are planning examples, not market averages.** A fast business decision can be triaged in a few days; customer discovery or a funding negotiation can take longer. Cost is the internal time actually used plus any approved specialist quotes; there is no one-size-fits-all budget.

## Stage 0 — Safety and authority gate (day 0)

Write: what decision is needed, by what date, whose legal authority is required, and what cannot be shared. If the matter affects securities, employment, fiduciary duties, litigation or a distressed company, route legal questions to qualified counsel immediately. The SEC's broker-dealer guidance makes clear why securities-related introductions and transaction compensation require separate scrutiny. [Broker-Dealers — US SEC](https://www.sec.gov/resources-small-businesses/capital-raising-building-blocks/broker-dealers)

**Deliverables:** decision owner; authority/approval map; confidentiality classification; specialist escalation list. **Stop condition:** no one can identify the lawful decision-maker or sensitive data would be exposed.

## Stage 1 — Frame the choice (day 1)

Write a five-sentence brief: *We need to decide [verb]; by [date]; because [event]; under [cash/legal/team constraints]; to preserve [objective].* Insist on at least three options, including the status quo, a narrower experiment, and a third path. Differentiate an objective from a predetermined answer; if the team already wants “hire a VP”, redefine the question as “how can we make sales repeatable?”.

**Deliverables:** scope statement, three-option matrix, explicit excluded choices. **Pitfall:** creating a straw-man alternative to legitimise the preferred decision.

## Stage 2 — Reconcile the baseline evidence (days 1–3)

Pull last three to six months of board materials, revenue and churn by cohort, customer interviews, hiring plans, payroll, cash movements, commitments, cap table, contracts and decision records. Use a redacted evidence register: ID, claim, source, date, reliability, owner and what contradicts it. Have finance verify the cash starting point. A separate runway model is useful because even sensible strategic options can be infeasible. Burkland's description of strategic-finance work gives examples of relevant inputs. [Burkland: Strategic Finance for Startups](https://burklandassociates.com/services/strategic-finance/)

**Deliverables:** one spreadsheet with a cash base case, downside case and option-specific cost; evidence register. **Costs:** internal analyst/finance hours; external modelling fees only after a scoped quote. **Pitfall:** confusing bookings or an optimistic pipeline with available cash.

## Stage 3 — Run buyer-side evidence tests (days 3–10, as feasible)

Create five hypotheses and five ways each could be wrong; conduct *suggested* 5–10 focused interviews or product observations, with the size adapted to the decision rather than sold as statistical proof. Prioritise recent buyers, non-buyers and churned users; ask about real behaviour and alternatives, not whether a speculative feature “sounds useful”. YC's interview guidance is a practical reference. [How To Talk To Users — Y Combinator](https://www.youtube.com/watch?v=z1iF1c8w5Lg)

**Deliverables:** interview guide, redacted notes, demand-versus-politeness map, observation log. **Pitfall:** selecting only fans or treating interview counts as representative market measurement.

## Stage 4 — Pre-mortem and dissent session (day 5–12)

Ask every decision participant to write down a private failure scenario before group discussion. Sort risks by consequence, warning signal, controllability and the evidence that would change their mind. Invite a colleague with no performance upside in the preferred answer. Rotate the person arguing for the status quo, pivot and reversible experiment.

| Failure statement | Early warning | Mitigation | Owner | Review |
|---|---|---|---|---|
| New market has no paying urgency | Buyers do not prioritise a paid pilot | Narrow customer segment, test payment | Commercial lead | Checkpoint date |
| Hiring burns runway before value | Time-to-productivity exceeds forecast | Stage hire, use interim support | CEO + finance | Monthly |
| Funding offer weakens future choices | Rights or dilution differ from expectations | Counsel review + scenario cap table | CEO + counsel | Before signing |

**Deliverables:** Pre-Mortem Workbook, 3 highest-severity risks, assigned actions. **Pitfall:** allowing the most senior participant to answer first.

## Stage 5 — Decision math and reversibility (day 7–14)

For each option show: cash spent before first evidence; time to first evidence; cash remaining if wrong; reversibility; people impact; legal dependency; and *what would make this option clearly inferior*. Keep probabilities as transparent subjective judgements when no data exist—never imply precision. For a SAFE or priced financing, run instrument-specific modelling and seek securities counsel; YC's SAFE materials describe why headline valuation alone is insufficient. [The SAFE — Y Combinator](https://www.ycombinator.com/safe)

**Deliverables:** one-page options grid with upside/downside and conditions. **Pitfall:** presenting a scenario spreadsheet as a forecast of guaranteed returns.

## Stage 6 — Decide, record dissent and communicate (day 10–15)

The CEO or designated authorised body makes the decision. Record the evidence available as of that date, strongest objection, conflicts, why each alternative was rejected and the earliest reliable signal that should reopen the choice. Send separate communications to team, board and outsiders according to confidentiality rules.

**Deliverables:** signed decision memo, authority record where needed, communications plan. **Pitfall:** retrospectively editing the reasoning to make a bad outcome seem inevitable.

## Stage 7 — Execute a bounded test (subsequent 2–6 weeks; example)

Move the first reversible commitment, not the full budget. Specify the actual output (qualified sales, paid tests, new operating cadence, legal-reviewed document), cash-at-risk ceiling and exact review meeting. YC's lecture on MVPs supports testing with simple iterations rather than polishing an unvalidated product. [How to Plan an MVP — Y Combinator](https://www.youtube.com/watch?v=1hHMwLxN6EM)

**Deliverables:** owner, dates, outcome dashboard, stop-loss criteria. **Pitfall:** resetting thresholds after results arrive.

## Stage 8 — Reopen or close the decision

Compare the observed evidence with the test specification, not merely with the team's mood. Decide to continue, modify, reverse or escalate; document what was learned even if the commercial outcome disappoints. Obtain independent outside review only for the *unanswered* issue and avoid retaining someone indefinitely without new decision work.

## Ready-to-use one-page decision memo

| Field | Entry |
|---|---|
| Decision / accountable owner / deadline | [ ] |
| Evidence for and against preferred option | [ ] |
| Three genuine alternatives | [ ] |
| Cash-at-risk and remaining downside runway | [ ] |
| Conflict and stakeholder map | [ ] |
| Specialist review required | [ ] |
| Minimum reversible action | [ ] |
| Measurable review / stop trigger | [ ] |

## When this becomes too much for DIY

Buy bounded expertise when there is no independent challenge inside the company, the deadline is too near to gather contradictory evidence, you have unresolved legal/security obligations, or cash models are not reconciled. A startup attorney handles legal interpretation; an independent challenger tests the decision; a fractional finance expert validates monetary constraints. Ask for a fixed scope, rights over inputs, data-handling arrangements and exact outputs—not a promise of certainty.

## Related founder FAQs

### How many customer interviews are enough?

There is no universal sample for a strategic choice; the suggested 5–10 interviews are a discovery starting point, not statistical proof. Check observed buying behaviour and recruit contradictory cases. [How To Talk To Users — Y Combinator](https://www.youtube.com/watch?v=z1iF1c8w5Lg)

### How much does the DIY process cost?

Count actual founder, finance, research and legal time and obtain written quotes for specialist work. No verified category median exists for an independent decision review.

### Should the board see the pre-mortem?

Share a decision-grade summary when appropriate and follow your governance duties and confidentiality rules. Do not automatically expose raw confidential interviews or private coaching material.

### What if the team refuses to challenge the CEO?

Collect independent written views before group discussion and record dissent neutrally. Consider a neutral facilitator whose fee is unrelated to the outcome.

### Can I use AI to draft the memo?

Yes, with redacted inputs and independent verification; label AI-generated hypotheses as unverified. NIST provides a voluntary GenAI risk framework for governance questions. [NIST Generative AI Profile, AI RMF](https://www.nist.gov/publications/artificial-intelligence-risk-management-framework-generative-artificial-intelligence)

### Who should own the final decision?

The authorised founder, CEO, board or other governance body owns it under the company’s arrangements. A paid adviser can recommend, but should not silently assume authority.

### What should trigger a restart?

Specify observable triggers in advance, such as cash below an agreed floor or failed product behaviour. Avoid retroactively moving thresholds to justify persistence.

## Sources

- [Broker-Dealers — US SEC](https://www.sec.gov/resources-small-businesses/capital-raising-building-blocks/broker-dealers)
- [Burkland: Strategic Finance for Startups](https://burklandassociates.com/services/strategic-finance/)
- [How To Talk To Users — Y Combinator](https://www.youtube.com/watch?v=z1iF1c8w5Lg)
- [The SAFE — Y Combinator](https://www.ycombinator.com/safe)
- [How to Plan an MVP — Y Combinator](https://www.youtube.com/watch?v=1hHMwLxN6EM)
- [NIST Generative AI Profile, AI RMF](https://www.nist.gov/publications/artificial-intelligence-risk-management-framework-generative-artificial-intelligence)

## Evidence gaps for the next research pass

- Calibrated timing and cost study from completed real founder decisions
- Permissioned example decision memos and counterfactual outcomes
- State-by-state governance and privilege checks for distressed companies

## FAQs
### How many customer interviews are enough?
There is no universal sample for a strategic choice; the suggested 5–10 interviews are a discovery starting point, not statistical proof. Check observed buying behaviour and recruit contradictory cases. [How To Talk To Users — Y Combinator](https://www.youtube.com/watch?v=z1iF1c8w5Lg)
### How much does the DIY process cost?
Count actual founder, finance, research and legal time and obtain written quotes for specialist work. No verified category median exists for an independent decision review.
### Should the board see the pre-mortem?
Share a decision-grade summary when appropriate and follow your governance duties and confidentiality rules. Do not automatically expose raw confidential interviews or private coaching material.
### What if the team refuses to challenge the CEO?
Collect independent written views before group discussion and record dissent neutrally. Consider a neutral facilitator whose fee is unrelated to the outcome.
### Can I use AI to draft the memo?
Yes, with redacted inputs and independent verification; label AI-generated hypotheses as unverified. NIST provides a voluntary GenAI risk framework for governance questions. [NIST Generative AI Profile, AI RMF](https://www.nist.gov/publications/artificial-intelligence-risk-management-framework-generative-artificial-intelligence)
### Who should own the final decision?
The authorised founder, CEO, board or other governance body owns it under the company’s arrangements. A paid adviser can recommend, but should not silently assume authority.
### What should trigger a restart?
Specify observable triggers in advance, such as cash below an agreed floor or failed product behaviour. Avoid retroactively moving thresholds to justify persistence.

## Sources
- [Broker-Dealers — US SEC](https://www.sec.gov/resources-small-businesses/capital-raising-building-blocks/broker-dealers)
- [Burkland: Strategic Finance for Startups](https://burklandassociates.com/services/strategic-finance/)
- [How To Talk To Users — Y Combinator](https://www.youtube.com/watch?v=z1iF1c8w5Lg)
- [The SAFE — Y Combinator](https://www.ycombinator.com/safe)
- [How to Plan an MVP — Y Combinator](https://www.youtube.com/watch?v=1hHMwLxN6EM)
- [NIST Generative AI Profile, AI RMF](https://www.nist.gov/publications/artificial-intelligence-risk-management-framework-generative-artificial-intelligence)