Finance & Analysis

Verified against ChatGPT · 2026-08-13

Build a valuation reasoning framework that compares methods instead of handing back a single number

Structures a comparison across valuation approaches for a specific situation, explaining what each method assumes and where they'd disagree, instead of producing one confident valuation figure that hides how much judgment went into it.

ChatGPT (GPT-5.1)4 fillable variables

The prompt

Ready to copy — highlighted parts are example details you can swap.

I need help structuring my thinking on a valuation question — not a single number, because a single confident valuation figure from an AI model would be misleading given how much judgment and market-specific data a real valuation actually requires. I want a framework comparing how different valuation approaches would reason about this situation and where they'd likely disagree.

SITUATION
Founder considering an early buyout offer for a 15%-owned stake in a private company

AVAILABLE FINANCIALS
Trailing 12-month revenue $3.2M, EBITDA roughly breakeven, no audited statements, two years of internal P&Ls

COMPARABLE CONTEXT I HAVE
One competitor raised at a reported 4x revenue multiple 18 months ago, but details of that deal's terms aren't public

WHAT THIS VALUATION IS FOR
Deciding whether to accept, negotiate, or reject a buyout offer for the stake

For each of the following approaches that's actually applicable to this situation — comparables/multiples, discounted cash flow, and (if relevant) an asset-based or precedent-transaction approach — explain what that method fundamentally assumes about value, what specific inputs it would need that I haven't given you (be concrete: for DCF, that means a discount rate and terminal growth assumption; for comparables, that means an actual peer set and their current multiples), and roughly what kind of situation makes that method more or less reliable. Then explain where these approaches would likely land in meaningfully different places for a situation like this one specifically, and why — not a generic "methods can differ" caveat, but the actual mechanism (e.g., a DCF is more sensitive to long-term growth assumptions than a multiples approach, which anchors to current market sentiment instead).

WHAT NOT TO DO
Do not produce a specific dollar valuation or multiple as if it were a real output — you do not have real comparable company data, a real discount rate derivation, or audited financials, and a specific number would carry false authority. Do not claim a named industry-standard multiple or discount rate as fact; if I haven't given you a number, say what's missing and what kind of source would supply it (an actual comparable transaction database, a real cost-of-capital calculation).

OUTPUT FORMAT
1. A section per applicable method: What It Assumes | What Inputs Are Missing | When This Method Is More/Less Reliable Here.
2. A short section on where these methods would likely disagree for this specific situation and why.
3. A closing paragraph stating plainly that this is a framework for organizing the analysis, not a valuation, and that an actual number requires either a qualified valuation professional or the real inputs (verified comparables, an actual discount rate, audited financials) plugged into a proper model.

Customize

Optional — swap in your own details for the highlighted parts above.

Why this works

Explicitly refusing to produce a single dollar figure or multiple, and instead structuring a comparison of what each method assumes, is the core design decision that keeps this prompt honest about what an AI model reasoning over incomplete inputs can actually contribute — a model asked directly "what is this worth" will produce a specific-sounding number by pattern-matching against typical multiples in its training data, and that number would carry an unearned appearance of rigor despite resting on no real comparable dataset, no derived discount rate, and no audited financials, which is precisely the kind of confident-but-baseless output this category exists to avoid. Requiring the model to name the concrete missing inputs for each method (an actual peer set with current multiples for comparables, a derived discount rate and terminal growth rate for DCF) rather than gesturing at methods abstractly is what makes this genuinely useful as a preparation exercise — it tells the user exactly what data they still need to gather before a real valuation could be responsibly attempted, which is actionable in a way a vague overview isn't. Explaining the actual mechanism behind why methods diverge (DCF's greater sensitivity to long-run growth assumptions versus comparables anchoring to current market sentiment) rather than a generic "different methods can give different answers" disclaimer gives the user a real intuition for which number to trust more in which market conditions, which is the actual reasoning skill this prompt is meant to build. The mandatory closing statement — that this organizes analysis but a real number needs a qualified valuation professional or actual verified inputs — keeps the tool correctly scoped as an explanation and structuring aid rather than something that could be mistaken for investment or valuation advice a reader might act on directly.

What you get back

Comparables approach assumes the market has already priced similar businesses correctly and that this company resembles that peer set closely enough to inherit its multiple — missing input: a real set of comparable transactions with disclosed terms, not a single rumored 4x figure from one competitor's raise. DCF assumes future cash flows can be reasonably projected and discounted at an appropriate rate reflecting risk — missing: an actual discount rate derivation and a defensible terminal growth assumption, which for a near-breakeven company is especially sensitive since small changes in the growth assumption swing the output disproportionately.

Verified against

ChatGPT GPT-5.1 · 2026-08-13

Changelog

  • 2026-08-13 Initial publish, verified against ChatGPT GPT-5.1.

Need this built into your business?

If a prompt isn't enough — what Scult builds, built and maintained for you — that's Scult's day job.

EXPLORE WHAT SCULT BUILDS
All Finance & Analysis prompts

Check your AI visibility

One URL in, a 0–100 score and the exact fixes out.

RUN THE CHECK

Browse all the tools

15 tools across six categories
13 of them never send your data anywhere

Free · No signup · No trial clock

SEE THE DIRECTORY