For CFOs and PE deal leads: build a full discounted cash flow model on your own numbers inside Claude. Revenue build, WACC, terminal value, sensitivity table, and football field, without an associate rebuilding it in Excel or a valuation shop fee.
Free — runs in your own ClaudeMedium setup · 4 steps12 ready-to-run prompts
Three minutes, four steps, nothing to install by hand
Claude sets it up for you. You just paste.
Never used Claude? It is free and takes 30 seconds to open. Copy the instruction below, paste it into Claude, and it reads this page and walks you through everything, one question at a time.
1
Tell Claude how to talk to you
One tap. It changes how much Claude explains, and how slowly it goes. You can change it any time.
2
Copy your setup instruction
A short instruction plus a link to this page lands on your clipboard. First copy asks for your email once. That unlocks every button across the whole library.
3
Open Claude in a new tab
Free account, no card, 30 seconds. This tab stays open so you can come back.
Claude reads this page, asks one question about your work, then guides you step by step until your first output is right. If anything looks wrong, tell Claude what you see, and it fixes it with you.
▸Prefer the full prompt instead of the link? (optional)
I am comfortable copy-pasting and following instructions, but I am not a developer.
There is nothing to install for this one and no commands to type: it all happens inside Claude. If any instruction below implies a Terminal, translate it into the equivalent click path for me instead.
- Plain English. Define jargon the first time it appears.
- One step at a time, then wait for me to confirm before the next one.
- Tell me what success looks like at each step, and diagnose any error before moving on.
Follow the instructions below with those rules applied.
I want to set up the DCF Model Builder inside my own Claude and run it on a real company this week. Walk me through it one step at a time, wait for me to confirm each step before the next, and define any term the first time you use it. Treat me like a CFO, PE deal lead, or analyst who values companies but has never wired up a Claude Project. This is NOT a Terminal install. There is nothing to install from a command line. It is all done by clicking inside Claude and pasting prompts. If any step needs a command window, you have misread this, so do not send me one.
Ask me this ONE question first, then stop and wait:
**Where does your company data live right now?**
- (A) In an Excel or Google Sheets model on my machine
- (B) In a PDF or set of filings (10-K, management accounts, a CIM)
- (C) In a data room or a system I would have to export from
- (D) Nowhere structured yet, I will paste raw numbers
Once I answer, walk me through the setup in this order, one step at a time:
1. **Open Claude and pin the model.** Tell me to go to claude.ai (or the desktop app), start a new chat, and set the model to Claude Opus 5 for the valuation math. Define: pinning the model means picking Opus 5 from the model dropdown so every prompt in this session uses it.
2. **Create a private Project.** Click "Projects" in the left sidebar, then "New project", name it something like "DCF — [company]". Explain why: a Project keeps my data in my own tenant. Nothing I load is uploaded to consultance.ai, stored by us, or seen by us. That is the whole reason this is safe to run on a live deal.
3. **Load my data the way that fits my answer above.**
- If (A): tell me to export the model to a clean sheet or upload the workbook into the Project knowledge, or use the Claude Excel add-in to read the sheet in place.
- If (B): upload the PDF or filings into the Project knowledge.
- If (C): tell me to export the specific statements I need first, then upload them. Do not tell me to connect a live system, that is implementation work, not part of this free setup.
- If (D): I will paste the figures straight into the chat when prompt 01 asks.
4. **Paste prompt 01, the setup router.** Tell me to open the vault, copy prompt 01, paste it into the Project chat, and answer its questions (valuation type, data source, my inputs). Explain that prompt 01 configures every later prompt, so it runs first, always.
Then run the first-session drill:
5. **Run the model on one real name.** Run prompts 02 through 09 in order. Do not skip. Point out that prompt 05 (WACC) and prompt 06 (terminal value) are where most of the answer is decided, so read those outputs carefully.
6. **Run prompt 10, the reconciliation, before I trust anything.** This re-derives the whole valuation a second, independent way and stops hard if the two do not tie within 2 percent. Tell me plainly: if it does not print RECONCILED, the base case is wrong, not the check. Nothing leaves my desk until it reconciles.
7. **What good output looks like.** Every figure traces to a source cell or a labeled assumption. The sensitivity grid shows a range, not a single number. The football field shows where the DCF overlaps comps, precedents, and the LBO. If a number appears with no source, that is the thing to fix before I show anyone.
House rules to state up front: I run prompt 01 first every time. I never strip a `<review_gate>` block, each one is my sign-off. I use a private Project so my data stays in my tenant. The official finance plugins and market-data connectors are a bonus path for later, never required to run this vault.
Step 2 · run it on your data
Step 1 set it up. These 12 prompts do the work.
the vault
The 12 prompts
Grab the whole pack as one file, or tap any prompt below to copy it on its own. Placeholders that look like {{THIS}} get swapped for your own numbers — and if you ran Step 1, Claude fills them in for you.
One .md file · all 12 prompts, numbered, in order · nothing left out.
<role>
You are a valuation desk lead. Convene a standing board of four named advisors and hold it for the whole session: a sell-side M&A associate who has built hundreds of DCFs under deadline, a buy-side investment committee member who kills weak assumptions for a living, a Big 4 valuation partner who signs fair-value opinions under ASC 820 and IFRS 13, and a CFO who has to defend the number to a board. Surface their disagreement, do not smooth it over.
</role>
<task>
Do NOT value anything yet. Set up the engagement. Ask me the blocks below, present lettered options, and WAIT for my answers before analysis.
1. VALUATION TYPE.
(A) Whole-company DCF for an M&A or IC decision
(B) Equity-research intrinsic value per share
(C) PE / LBO entry valuation to triangulate against a return model
(D) Fairness-opinion grade valuation for an audit or board gate
2. DATA SOURCE. This decides how every later prompt runs.
(A) I will upload financials into this Claude Project's knowledge
(B) I will paste raw figures into the chat
(C) I will use the Claude app add-in to read an Excel workbook open on my machine
(D) I will connect a governed data room or warehouse connector
(E) A mix
3. INPUTS. Capture these tokens now, flag any I leave blank:
{{COMPANY}} {{CURRENCY}} {{FISCAL_YEAR_END}} {{YEARS_OF_HISTORY}} {{FORECAST_YEARS}}
{{TAX_RATE}} {{RISK_FREE_RATE}} {{EQUITY_RISK_PREMIUM}} {{BETA}} {{PRE_TAX_COST_OF_DEBT}}
{{TARGET_DEBT_WEIGHT}} {{TERMINAL_GROWTH}} {{TERMINAL_EXIT_MULTIPLE}} {{NET_DEBT}} {{SHARES_OUT}}
4. OUTPUT BAR. Confirm for the session: every figure traces to a source or a labeled assumption, every assumption is labeled with who it came from, and where you estimate you say so and give the basis.
</task>
<output_format>A short setup confirmation: valuation type, data source, the token table with blanks flagged, the output bar in one line. Then stop and wait for me to say "build".</output_format>
<role>You are the associate cleaning the historicals before anyone models off them.</role>
<task>Work from the data source I selected in prompt 01. Pull {{YEARS_OF_HISTORY}} years of income statement, balance sheet, and cash flow for {{COMPANY}}. Strip one-off items and label each adjustment. Restate revenue, EBITDA, EBIT, D&A, capex, and change in net working capital on a clean basis. Compute the historical margins, capex percent of revenue, and NWC percent of revenue that will anchor the forecast.</task>
<constraints>Use only the data source from prompt 01. Every normalization gets a one-line reason and a source. Flag any inferred line.</constraints>
<output_format>A clean historical table, a margins-and-ratios table, and a list of every adjustment with reason and source.</output_format>
<review_gate>End: "Confirm these normalized historicals before I build the forecast."</review_gate>
<role>You are the associate building revenue bottom up, not by a single growth rate.</role>
<task>Project revenue for {{FORECAST_YEARS}} years using explicit drivers. Pick the structure that fits {{COMPANY}}: volume times price, customers times ARPU times retention, capacity times utilization, or units times ASP. Show each driver's path and the assumption behind it. Tie year one to the last historical year from prompt 02.</task>
<constraints>Use the data source from prompt 01. Label every driver assumption and its basis. Flag any year where growth exceeds the historical peak and justify it.</constraints>
<output_format>A driver table by year, the resulting revenue line, and one paragraph on what has to be true for this to hold.</output_format>
<review_gate>End: "These are the growth drivers a buyer attacks first. Which one do you want to stress?"</review_gate>
<role>You are the associate turning revenue into unlevered free cash flow.</role>
<task>From prompt 03, project to unlevered FCF each year: revenue to EBIT via the margin path from prompt 02, less cash taxes at {{TAX_RATE}}, plus D&A, less capex, less change in net working capital. Hold margins, capex percent, and NWC percent to the prompt 02 anchors unless I override, and flag every override.</task>
<constraints>Use the data source from prompt 01 and the outputs of 02 and 03. Unlevered basis only. Show the full walk.</constraints>
<output_format>A year by year FCF bridge from revenue to unlevered FCF, every intermediate line visible.</output_format>
<review_gate>End: "Confirm the FCF conversion. Does the capex and working-capital path match how this business funds growth?"</review_gate>
<role>You are the valuation partner who has to defend the discount rate to an auditor.</role>
<task>Build WACC for {{COMPANY}}. Cost of equity via CAPM: {{RISK_FREE_RATE}} plus {{BETA}} times {{EQUITY_RISK_PREMIUM}}, adding a size or country premium only if I ask and labeling it. After-tax cost of debt: {{PRE_TAX_COST_OF_DEBT}} times (1 minus {{TAX_RATE}}). Weight by {{TARGET_DEBT_WEIGHT}} target structure. Show every input, its source, and the final WACC, and name the input the answer is most sensitive to.</task>
<constraints>Use the inputs from prompt 01. If any is blank, propose a defensible market figure, label it ASSUMPTION, and cite the basis. Do not bury a plug.</constraints>
<output_format>The CAPM build, the cost-of-debt build, the weighting, the single WACC, and "the number most likely to be challenged here is ___."</output_format>
<review_gate>End: "This rate moves the valuation more than anything. Sign off on the WACC or tell me which input to change."</review_gate>
<role>You are the associate who knows terminal value quietly drives most of the answer.</role>
<task>Compute terminal value two independent ways and reconcile: Gordon growth (final-year FCF times (1 plus {{TERMINAL_GROWTH}}) divided by (WACC minus {{TERMINAL_GROWTH}})) and exit multiple (terminal EBITDA times {{TERMINAL_EXIT_MULTIPLE}}). Show the implied growth the multiple bakes in and the implied multiple the growth bakes in. Flag if they disagree by more than 15 percent and say which you trust.</task>
<constraints>Use outputs from 04 and 05. {{TERMINAL_GROWTH}} must not exceed long-run nominal GDP unless justified. State the percent of enterprise value in the terminal; if above 75 percent, say so loudly.</constraints>
<output_format>Both calculations, the cross-checks, the percent of EV in the terminal, and your recommendation.</output_format>
<review_gate>End: "Terminal value is ___ percent of this valuation. Confirm the method before I discount."</review_gate>
<role>You are the associate assembling enterprise and equity value.</role>
<task>Discount every forecast-year FCF (prompt 04) and the terminal value (prompt 06) to present value at the WACC (prompt 05), mid-year convention unless I say otherwise. Sum to enterprise value. Bridge to equity: less {{NET_DEBT}}, plus non-operating assets. Divide by {{SHARES_OUT}} for value per share. Show the discount factor per year.</task>
<constraints>Use only the outputs of 04, 05, 06 and the tokens from 01. Show the full waterfall, no step hidden.</constraints>
<output_format>A PV table with discount factors, the EV total, the EV-to-equity bridge, and value per share. State the headline EV in one line.</output_format>
<review_gate>End: "This is the base case. Do not circulate until the sensitivity and reconciliation prompts have run."</review_gate>
<role>You are the associate who never shows a point estimate without a range.</role>
<task>Build two 5 by 5 grids around the base case from prompt 07: enterprise value across WACC (plus and minus 1.0 percent, 0.5 steps) by terminal growth (plus and minus 1.0 percent, 0.5 steps), and EV across WACC by exit multiple. Mark the base-case cell. Call out the swing from low corner to high corner as a percent of base.</task>
<constraints>Use the model from prompt 07. Recompute each cell fully. Keep every other assumption fixed. Label axes and the base cell.</constraints>
<output_format>Two labeled grids, base cell marked, and one line on total valuation swing.</output_format>
<review_gate>End: "This range, not the point estimate, goes to committee. Confirm the bounds they will ask about."</review_gate>
<role>You are the associate who never lets a DCF stand alone.</role>
<task>Place the DCF range from prompt 08 next to trading comps (EV/EBITDA and EV/revenue from a peer set), precedent transactions (multiples paid in comparable deals), and an LBO (the entry EV that clears my return hurdle, if I give it). Show each as a low-to-high bar, mark the overlap, and state the defensible range where the methods agree.</task>
<constraints>Use the data source from prompt 01 for peers and deals. Do not invent comps or deal multiples. If I have not supplied them, ask before drawing the bar.</constraints>
<output_format>A football-field layout, one bar per method, the overlap zone, and the concluded range in one line.</output_format>
<review_gate>End: "Where the four methods overlap is your defensible range. Confirm the peer set and deal set."</review_gate>
<role>You are a second valuation reviewer who did not build this model and does not trust it yet.</role>
<task>Re-derive enterprise value a SECOND, independent way and reconcile to prompt 07. Rebuild unlevered FCF from the raw historicals and driver assumptions without reusing prompt 04's cells, apply WACC and terminal value independently, and compare. If the two differ by more than 2 percent, STOP: list every diverging line, largest gap first, with the likely cause. Only when they tie within 2 percent, output RECONCILED and the agreed enterprise value.</task>
<constraints>Work only from the raw data source (prompt 01) and the labeled assumptions. Do not copy forward prompt 04 or 07 outputs. No RECONCILED unless the numbers actually tie.</constraints>
<output_format>Either a divergence list ordered by size with causes, or "RECONCILED — EV = ___" with a one-line tie-out.</output_format>
<review_gate>End: "Nothing circulates until this prints RECONCILED."</review_gate>
<role>You are the buy-side IC member trying to kill this valuation before committing capital.</role>
<task>Attack the model from prompts 03 to 10: which growth driver is doing the most work and what a 20 percent softer read does, whether the WACC is generous and what a harsher rate does, whether terminal value carries too much, and the single assumption that breaks the thesis if wrong. For each: the challenge, the value impact, the strongest honest rebuttal.</task>
<constraints>Use the built model. No untraceable numbers. Be adversarial. If an assumption is indefensible, say so; do not manufacture a rebuttal.</constraints>
<output_format>A challenge table: assumption, attack, value impact, honest rebuttal. Then the one assumption you would refuse to sign.</output_format>
<review_gate>End: "These are the questions you will get. Which one do you not yet have a clean answer to?"</review_gate>
<role>You are the associate packaging the model for the committee.</role>
<task>Assemble a self-contained valuation summary for {{COMPANY}}: the concluded EV and range (08 and 09) gated on the RECONCILED flag from 10, the base-case DCF walk in brief, the sensitivity range and football-field overlap, the three assumptions the value hinges on with basis, and the open questions from prompt 11.</task>
<constraints>Every figure traces to a prompt output and through it to a source or labeled assumption. Do not restate a number the reconciliation did not clear.</constraints>
<output_format>A one-page IC summary: headline range, base case, sensitivity, football field, key assumptions with basis, open questions. Auditor-ready.</output_format>
<review_gate>End: "This is what leaves your desk. Confirm every number ties to a source before committee."</review_gate>
Got the prompts. Want them wired into your actual stack? We map that on a free AI audit.
• Run last quarter's numbers first. Live data is not a test bed.
• Nothing here uploads to us. It runs in your own Claude account, on your own machine.
• A named human reviews and signs every output before it reaches a board, lender, or client.
• Mask account numbers and names to the minimum the task needs.
the fine print
Straight answers on ownership
Prompt set authored by consultance.ai. A DCF built here is a first-draft valuation, not a fairness opinion or investment advice. Your data stays in your own Claude tenant; we never see it. Fair-value work maps to ASC 820 and IFRS 13, and valuation performance to AICPA SSVS No. 1 and IVS 105; a named human signs off every figure against source before it reaches an investment committee, a board, or an auditor. Not legal or investment advice.
Want this running in your business, not just your laptop? We build it and hand you the keys.
Free DCF Model Builder is a finance and data build in the consultance.ai AI Build Library. For CFOs and PE deal leads: build a full discounted cash flow model on your own numbers inside Claude. Revenue build, WACC, terminal value, sensitivity table, and football field, without an associate rebuilding it in Excel or a valuation shop fee. It fits CFOs, PE deal leads and principals, investment committee members, corp dev, and owner CEOs who have to sign off a valuation. Setup difficulty is Medium, with 4 plain-English steps.
What does Free DCF Model Builder do?
For CFOs and PE deal leads: build a full discounted cash flow model on your own numbers inside Claude. Revenue build, WACC, terminal value, sensitivity table, and football field, without an associate rebuilding it in Excel or a valuation shop fee.
Who is Free DCF Model Builder for?
It fits CFOs, PE deal leads and principals, investment committee members, corp dev, and owner CEOs who have to sign off a valuation.
How hard is Free DCF Model Builder to set up?
Medium to set up — one guided setup instruction covering 4 plain-English steps, plus 12 ready-to-run prompts on the resource page.
How would consultance.ai build this out?
We build the DCF workflow into your own environment, your deal data never leaves the building: the prompts loaded into your Claude tenant, a connector to your model and data room, the reconciliation gate wired as a standing pre-distribution control, and a retained assumption log so the valuation stands up in an audit. Done with you, then handed over so you own it.
What are the licensing terms?
Prompt set authored by consultance.ai. A DCF built here is a first-draft valuation, not a fairness opinion or investment advice. Your data stays in your own Claude tenant; we never see it. Fair-value work maps to ASC 820 and IFRS 13, and valuation performance to AICPA SSVS No. 1 and IVS 105; a named human signs off every figure against source before it reaches an investment committee, a board, or an auditor. Not legal or investment advice.
Want this built into your workflow?
Free DCF Model Builder is the starting point. On a free AI audit we map where it fits your stack and what consultance.ai would build around it.