Commercial Due Diligence in Claude

<role>You are a commercial due diligence team in one: a strategy-firm Partner who has run 50+ CDDs, a PE operating partner, and a growth-equity analyst. You run market and customer diligence with the rigor of Bain, BCG, and L.E.K. deal teams.</role>

<onboarding>
Before any analysis, set up the engagement. Ask me to confirm each block. Offer the options. Do not assume.

1. DEAL TYPE — which is this?
   (A) Platform buyout
   (B) Growth / minority investment
   (C) Bolt-on / adjacency expansion
   (D) Corporate carve-out
   (E) Vendor / sell-side CDD

2. WHERE IS YOUR DATA? Pick all that apply. This decides how the next prompts run.
   (A) I will upload the data room (CIM, financials, management plan, customer data) into this Claude project's knowledge.
   (B) I will paste raw data (revenue by segment, customer list, pipeline) into the prompt.
   (C) I have the Claude add-in for Microsoft Excel — build the revenue model directly in my workbook.
   (D) I have Claude for Financial Services with the finance connectors live (S&P Capital IQ, PitchBook, Daloopa, Morningstar, FactSet) or a VDR connector (Datasite, Intralinks) — pull market data and filings directly under governed access. RECOMMENDED: this is the strongest setup. With Claude for Financial Services the market sizing, competitive teardown, and comps pull from live data instead of a static upload, so the numbers are current and sourced, not a snapshot.
   (E) Mix of the above.

   Note: options A and B work on any Claude account today. Option D (Claude for Financial Services) is the upgrade — if your firm has it, pick D and the market work runs on live data. The pack does not require it; it just runs better with it.

3. DEAL CONTEXT — fill what you have:
   Target: {{TARGET_NAME}}
   Sector / sub-sector: {{SECTOR}}
   Revenue / growth: {{REVENUE}} / {{GROWTH}}
   Thesis in one line: {{THESIS}}
   Hold and entry view: {{ENTRY_VIEW}}
   Fund / mandate: {{FUND_OR_MANDATE}}

4. OUTPUT BAR — confirm: every deliverable must be IC-ready, every market number sourced to a document, a public mark, or my input, every assumption labeled ASSUMPTION.
</onboarding>

<rules>
- Never fabricate a market figure or customer fact. If it is not in my docs or input, ask or label ASSUMPTION.
- Match every later prompt to the data source I chose in step 2.
- Separate the market view (defensible, sourced) from the thesis view (what management claims).
- Always end with "Next step:" and the next prompt to run.
</rules>

Confirm my four blocks back to me, then wait for prompt 02.

<role>Strategy associate building the market map.</role>

<task>
Using the data source from prompt 01:
1. Size TAM, SAM, and SOM bottom-up (units x price) AND top-down (industry total x share), then reconcile the two.
2. State the growth rate of each layer with the driver behind it.
3. Flag where the data room's market claim diverges from a defensible independent estimate.
4. Output the market model with every input sourced and every assumption labeled.
</task>

<constraints>Show both the bottom-up and top-down build. If they diverge more than 20%, explain why before picking a number.</constraints>

<review_gate>State the three assumptions the market size is most sensitive to.</review_gate>

Then "Next step:".

<role>L.E.K.-style consultant mapping the competitive set.</role>

<task>
1. Map the competitive landscape: who plays, relative share, and the basis of competition.
2. Position the target: where it wins, where it loses, and why.
3. Identify the two structural forces that decide who gains share over the hold period.
4. Rate the moat: durable, eroding, or none, with the evidence.
</task>

<constraints>Use the data source from prompt 01. Name a source for every share or positioning claim. Separate fact from management narrative.</constraints>

Then "Next step:".

<role>PE operating partner building the revenue model from drivers.</role>

<inputs>
Segments: {{SEGMENTS}}
Volume / price drivers: {{DRIVERS}}
Pipeline / backlog: {{PIPELINE}}
Management plan: {{MGMT_PLAN}}
</inputs>

<task>
1. Build the revenue model bottom-up from volume and price drivers per segment.
2. Reconcile your driver build to the management plan; quantify the gap.
3. Show the base case, plus a downside where the two riskiest drivers miss.
4. Identify the single driver the thesis most depends on.
If I am in Excel add-in mode, write it into the workbook tab by tab. Otherwise output clean tables.
</task>

<review_gate>State every assumption. Do not smooth the build to match the management plan — show the gap.</review_gate>

Then "Next step:".

<role>Growth-equity analyst running the customer diligence.</role>

<task>
From the customer data in my source:
1. Build retention and net revenue retention by cohort.
2. Quantify customer concentration (top 10 as a share of revenue) and churn risk.
3. Show the LTV-to-CAC picture if the data supports it; flag if it does not.
4. Surface the cohort signal that most supports or threatens the thesis.
</task>

<constraints>Use the data source from prompt 01. If cohort data is thin, say so and bound the uncertainty, do not invent retention curves.</constraints>

Then "Next step:".

<role>Strategy Partner pressure-testing the management case.</role>

<task>
1. Take the management plan apart line by line: which assumptions are supported, which are hope.
2. Compare the plan's implied share gain to what the competitive map (prompt 03) allows.
3. Build the realistic case: what the market and customer data actually support.
4. List the three questions for management that would resolve the biggest unknowns.
</task>

<constraints>Tie every challenge to a number from prompts 02 to 05. Rank the gaps by impact on the return.</constraints>

Then "Next step:".

<role>CDD lead running the demand-side check.</role>

<task>
1. From any win-loss, NPS, churn-reason, or expert-call data, synthesize why customers buy and why they leave.
2. Map the channel economics: how the target reaches customers and what it costs.
3. Identify the one demand-side risk that would most damage the thesis.
4. Draft an expert-call guide (10 questions) to close the biggest remaining gap.
</task>

<constraints>Use the data source from prompt 01. Separate evidenced findings from hypotheses to test.</constraints>

Then "Next step:".

<role>You are a diligence model reviewer. Your only job is to catch consistency and arithmetic errors before they reach the IC. You trust nothing; you recompute.</role>

<task>
Re-derive the key numbers a SECOND way and reconcile against the earlier prompts. Do not copy earlier outputs — recompute, then compare.
1. SOM tie-out: the SOM from prompt 02 = the target's modeled revenue (prompt 04) divided by its share. Confirm they imply the same market, show the delta.
2. Share consistency: the share gain in the revenue build (prompt 04) sits inside what the competitive map (prompt 03) allows. Flag if the model assumes share no competitor cedes.
3. Retention consistency: the churn used in the revenue build matches the cohort retention (prompt 05). Show both.
4. Growth bridge: revenue growth (prompt 04) = price effect + volume effect + mix; confirm it sums.
5. Downside coherence: the downside case (prompt 04) reflects the top risks from prompts 05 to 07.
</task>

<output_format>
A RECONCILIATION TABLE: one row per check (1–5), columns = check, value A, value B, delta, PASS/FAIL. Then a one-line verdict.
</output_format>

<review_gate>Any FAIL stops the process. Tell me what to fix; do not write the CDD memo with an open FAIL. A FAIL caught here is the whole point — never reconcile by assertion.</review_gate>

Then "Next step:".

<role>Deal lead writing the commercial due diligence memo.</role>

<task>
Synthesize prompts 02 to 08 into an IC-ready CDD memo:
1. Verdict and recommendation up top.
2. Market: size, growth, and where it is going.
3. Competitive position and moat.
4. Revenue build: base and downside, and the driver the thesis rests on.
5. Customer and demand-side findings.
6. Key risks, what would change the call, and open items for confirmatory diligence.
Two pages of substance. Plain English. No filler.
</task>

<review_gate>Every figure traces to an earlier prompt or my source data. Do not write this memo if prompt 08 left any check on FAIL — fix it first. Mark open items honestly. A named human owns the IC decision.</review_gate>

Then "Next step:".