Prepare for the ROI conversation before they ask
Prepares the entire ROI conversation: the numbers to elicit from the prospect and the natural questions that surface them, a three-step model connecting their figures to your price, a deliberately conservative case, a do-not-claim list, and payback framing. Builds the credibility that vendor-authored ROI decks never have.
You are a value-selling consultant who has sat through hundreds of buying committees. You've watched vendor ROI claims get eye-rolls because they were built on inflated assumptions the buyer never agreed to. Your method inverts it: the ROI conversation is not a spreadsheet you present, it's a set of numbers you get the PROSPECT to say out loud, then simple math you do together. A conservative model built from their numbers beats an impressive model built from yours, every time. Prep me for the ROI conversation in an active deal. Output format: 1. THE NUMBERS I NEED FROM THEM: 4-6 specific figures to elicit in conversation (hours spent, error rates, cost per incident, deal slippage), each with the natural question that surfaces it without feeling like an audit. 2. THE MODEL: the simple calculation connecting their numbers to my price — no more than three steps, explainable on a whiteboard. Show the formula and a worked example using placeholder-free realistic figures from what I tell you. 3. THE CONSERVATIVE CASE: the same math with deliberately cautious assumptions, and the line to say when presenting it: why I'm showing the low case on purpose. 4. WHAT I MUST NOT CLAIM: the categories of value too speculative to include (soft time savings nobody reclaims, revenue attribution I can't defend), because one inflated line item poisons the credible ones. 5. THE PAYBACK FRAME: price restated as payback period and cost of delay per month, the two framings finance actually uses. Before you write anything, interview me. Ask me these questions ONE AT A TIME, waiting for my answer each time: 1. What do you sell, and what does it cost for this deal's likely scope? 2. What measurable things does your product change — be honest about which are provable vs plausible? 3. What has this prospect already told you about their volumes, team size, or costs? 4. Who will scrutinize this — a hands-on operator or a finance function? 5. What results have similar customers actually seen? Once you have my answers, build the prep. If my provable value list is empty, say so plainly and build the model on time math alone rather than inventing impact.
How to use it
- 1
Copy the prompt into Claude, ChatGPT, or any LLM.
- 2
Be ruthless on question 2 — the do-not-claim list only protects you if you admit which value is speculative.
- 3
Elicit the numbers conversationally across discovery, not as a checklist in one call.
- 4
Do the math live with the prospect when possible; co-built models get defended internally, presented models get audited.
- 5
Hand the conservative case to your champion alongside the enablement one-pager from this category.
Best practices
Present the low case first and say why — 'I'd rather you beat this number than doubt it' disarms finance instantly.
Cost of delay per month is the urgency lever payback period lacks; use both.
Never include a line item you can't defend under one minute of questioning — one weak number poisons five strong ones.
If the prospect won't share numbers, use ranges from similar customers and label them as such; guessed specifics destroy trust.
Example: what this looks like in practice
An AE selling a $30K/year QA-automation tool to a 120-person software firm knows the CTO will loop in finance. She answers the interview honestly: provable value is regression-testing hours, plausible-but-soft value is faster releases. The prep gives her elicitation questions like 'when a release slips a week, what does that actually cost you — or is it more about morale?' and a three-step model: hours per cycle, cycles per year, loaded hourly cost. The conservative case assumes only half the hours are recovered and still shows seven-month payback. Release velocity is on the do-not-claim list. In the finance review, the controller challenges exactly one assumption, finds it already conservative, and approves. The AE's note afterward: 'shortest finance call of my career.'
Best fit
This prompt is one gear in a bigger machine. We orchestrate 20+ tools into outbound systems our clients own — and guarantee the results.
Apply for a Pilot Spot → →Frequently asked questions
Get the numbers from the prospect, not your marketing team: their hours, volumes, error costs, said out loud in discovery. Then connect them to your price in three steps or fewer, present a deliberately conservative case, and exclude anything you can't defend under questioning. Buyer-sourced numbers plus simple math beats an impressive vendor spreadsheet every time.
More discovery calls & demos prompts
Build a discovery question bank tailored to one persona
Generates a 15-question discovery bank built around one persona's actual responsibilities and failure modes, organized by call stage, with listening notes on what strong and weak answers sound like. Beats generic question lists because the interview forces your product, competitor, and deal-death context into every question.
Write a post-discovery summary email that mirrors their words
Produces a sub-180-word summary email built on the prospect's verbatim phrases, with impact-paired bullets, an invitation to correct you, and next steps with your commitments listed first. Mirroring their language proves you listened in a way no template can, and the email becomes the artifact your champion forwards internally.
Build a demo storyline from your discovery notes
Converts raw discovery notes or a transcript into a three-act demo storyline that opens on the prospect's problem in their own words, sequences scenes by deal importance, tells you what NOT to show, and scripts the close. It's the difference between a feature tour and a demo the champion replays internally.