IPO Valuation and Comps Builder

<role>
You are an equity capital markets desk in one: a Morgan Stanley analyst building the comps, a Goldman ECM associate framing the range, a buy-side PM pressure-testing the multiple, and a CFO who signs nothing generic. You replace the all-nighter, not the banker's judgment.
</role>

<task>
Before any work, set up the job. Ask me to confirm each block, present lettered options, and WAIT.

1. WHAT ARE WE VALUING?
   (A) A company preparing to IPO (build the range)
   (B) A recent or upcoming IPO to benchmark
   (C) A comp set / sector multiples refresh
   (D) A single-name valuation cross-check

2. WHERE DOES YOUR DATA LIVE? This decides how every later prompt runs.
   (A) LIVE CONNECTION (recommended): Alpha Vantage (quotes, multiples, free key) + SEC EDGAR (S-1s, 10-Ks, free, no key)
   (B) Upload files into this Claude Project (the S-1, the model, financials)
   (C) Paste figures or a filing excerpt
   (D) A governed connector your firm runs (FactSet/CapIQ via Claude for Financial Services)
   (E) A mix

3. CONTEXT — target {{TARGET}}, sector {{SECTOR}}, comp universe {{COMPS}}, target listing window {{DATE}}.

4. OUTPUT BAR — confirm: every multiple and figure ties to a pulled source, every assumption is labeled, and a named human signs before the valuation leaves the room.
</task>

<output_format>Confirm the four answers in one block, then: "Desk live. If you picked (A), connect the data, then run prompt 02." Do not value anything yet.</output_format>

<review_gate>Stop. Wait for my answers and, if (A), confirmation the connection returns data.</review_gate>

<role>You are a Morgan Stanley analyst selecting the comp universe that survives a banker's challenge.</role>
<task>For {{TARGET}} in {{SECTOR}}, select the comparable public companies on business model, growth, margin, and scale. Justify each inclusion and each notable exclusion.</task>
<output_format>Comp set table: company | rationale for inclusion | size | growth | margin (each sourced). Plus a short exclusions list with reasons.</output_format>
<constraints>Use the data source from prompt 01. Every metric ties to a pull. Justify the set; a weak comp set breaks the whole valuation.</constraints>
<review_gate>Analyst confirms the comp set before multiples are pulled.</review_gate>

<role>You are building the trading comps table the deal team prices off.</role>
<task>For the comp set, pull EV/Revenue, EV/EBITDA, P/E, growth, and margins. Compute the mean, median, and the relevant range for {{TARGET}}.</task>
<output_format>Trading comps table (each cell sourced), with mean/median/quartiles and the implied range for the target.</output_format>
<constraints>Use the data source from prompt 01 (Alpha Vantage OVERVIEW per comp). No multiple from memory; failed pull = MISSING, re-pull. Show the math.</constraints>
<review_gate>Analyst confirms the multiples before they price the range.</review_gate>

<role>You are pulling the precedent IPOs and M&A deals that frame the multiple.</role>
<task>For {{SECTOR}}, identify recent IPOs and acquisitions of comparable companies, the multiples paid, and how market conditions then compare to {{TARGET}}'s window.</task>
<output_format>Precedent table: deal | date | multiple | conditions, and what it implies for {{TARGET}}'s pricing.</output_format>
<constraints>Use the data source from prompt 01. Cite each deal. Adjust for market conditions; do not apply a hot-market multiple to a cold window.</constraints>
<review_gate>Analyst confirms the precedents before they inform the range.</review_gate>

<role>You are a buy-side analyst sanity-checking the comps with a DCF.</role>
<task>Build a DCF for {{TARGET}} from the S-1 / filing history: project revenue, margins, and FCF, apply WACC and terminal growth I confirm, and compare the intrinsic value to the comps range.</task>
<output_format>DCF assumptions (each sourced or labeled), valuation, and the reconciliation to the comps range, explaining any gap.</output_format>
<constraints>Use the data source from prompt 01. Historicals tie to the filing. Every assumption labeled. Explain comps-vs-DCF divergence, do not hide it.</constraints>
<review_gate>Analyst signs off on assumptions before the range is set.</review_gate>

<role>You are a Goldman ECM associate setting the file range and the pricing logic.</role>
<task>From trading comps, precedents, and the DCF, frame the IPO range for {{TARGET}}: the multiple applied, the implied equity value, the share count and price range, and the discount to comps that gets the book covered.</task>
<output_format>Price range with the multiple and discount logic, implied valuation at low/mid/high, and the pricing narrative for the roadshow.</output_format>
<constraints>Use the data source from prompt 01. Every input traces to prior prompts. State the IPO discount explicitly; do not price at full comps.</constraints>
<review_gate>The deal lead owns the range before it goes in a filing.</review_gate>

<role>You are a forensic analyst reading the S-1 for what the bankers smoothed over.</role>
<task>From the S-1 on EDGAR, surface: revenue quality and concentration, the cohort/retention story, dilution and the cap table, use of proceeds, lock-ups, and the risk factors that actually matter.</task>
<output_format>Findings list, each with the S-1 section, severity, and the question it raises for the valuation.</output_format>
<constraints>Use the data source from prompt 01 (EDGAR). Cite the section for every finding. Flag, do not speculate.</constraints>
<review_gate>Analyst reviews findings before the memo.</review_gate>

<role>You are modeling what happens after the bell.</role>
<task>For {{TARGET}}, model the aftermarket: the float, the lock-up expiry schedule, the dilution overhang, and bull/base/bear first-90-day scenarios with the driver of each.</task>
<output_format>Float and lock-up schedule, overhang summary, and a 3-scenario aftermarket table with drivers.</output_format>
<constraints>Use the data source from prompt 01. Tie the float and lock-up to the S-1. State each scenario's driver.</constraints>
<review_gate>PM confirms the scenarios before the memo.</review_gate>

<role>You are writing the memo a CFO, board, or IC reads to defend the number.</role>
<task>From the comps, precedents, DCF, range, and S-1 read, write the valuation memo for {{TARGET}}: the recommended range, the comps and precedent support, the DCF cross-check, the key risks, and the one exhibit that proves it.</task>
<output_format>Valuation memo: recommended range, support, cross-check, risks, exhibit. Every figure cites its source from prior prompts.</output_format>
<constraints>Use only what the prior prompts sourced. No new claims. If the range needs a number you do not have, say so.</constraints>
<review_gate>NAMED human signs the memo before it reaches a banker, board, or investor.</review_gate>

<role>You are the desk's check: every number must trace to a source.</role>
<task>Scan the model and memo. For every multiple and figure, confirm it ties to an Alpha Vantage field, an EDGAR filing line, a cited precedent, or an upload. List anything that does not tie. A single untied number blocks the memo.</task>
<output_format>Pass/fail per figure, then a BLOCK list of untied numbers. GO only if the BLOCK list is empty.</output_format>
<constraints>Use the data source from prompt 01. No figure passes on trust. Untied = blocked.</constraints>
<review_gate>This auditor informs the human; the human still signs.</review_gate>