Verified against ChatGPT · 2026-08-03
Stress-test a deal before you call it committed on the forecast
Runs a specific deal's signals against the documented ways B2B deals actually slip, so real risk gets flagged before a forecast call instead of after the deal slides another quarter, with one reason driving the risk level, not an averaged score.
The prompt
Ready to copy — highlighted parts are example details you can swap.
I need an honest risk read on this deal before I call it commit on our forecast.
Deal: Vantage Freight Co., $68,000 ACV, expected close August 29, 2026.
Last contact: 11 days ago with {{last_contact_person}}.
Stakeholders engaged so far: one ops manager; no one from finance or leadership has joined a call.
Economic buyer confirmed: named but never joined a call, all info is secondhand from the ops manager.
Next step booked: no, they said they would "circle back after budget review".
Anything that's changed or gone quiet recently: {{recent_changes}}.
Assess this deal against these documented ways deals slip, and for each, tell me plainly whether it applies here and how strongly, with one supporting sentence grounded in the actual details above, not a generic statement that would apply to any deal:
1. SINGLE-THREADED — only one stakeholder engaged, no direct relationship with the actual economic buyer
2. NO CONFIRMED NEXT STEP — momentum described in words but nothing calendared with a date
3. VERBAL YES, NO PROCESS — enthusiasm expressed without a confirmed decision process, approval chain, or timeline behind it
4. GOING QUIET — a drop in responsiveness that doesn't match the urgency the deal supposedly has
5. COMPETITIVE OR STATUS-QUO RISK — a real alternative, including "do nothing," that hasn't actually been ruled out
OVERALL RISK LEVEL
Give me an overall risk level — Low, Medium, or High — with the ONE reason actually driving that level, not an average across all five categories. A deal that's High risk because of one severe factor and Low risk on the other four is still High risk overall; averaging the five into a comfortable Medium hides exactly the thing that matters most.
SINGLE NEXT ACTION
Give me the single next action that would most reduce risk — not a list of five things to do, since a list of five equally-weighted actions is how a real risk gets buried among lower-priority housekeeping tasks and nothing actually gets fixed before the next forecast call.
WILLINGNESS TO DOWNGRADE
Be willing to tell me this deal should move to a later quarter or be marked at-risk even if that's not what I want to hear going into the forecast call — a forecast that already has this deal counted as commit creates real pressure to find reasons it's fine, and the read here needs to hold up against that pressure, not accommodate it.
FORMAT
A short table for the five risk categories (category, applies?, strength, supporting detail), followed by the overall risk level and reason, the single next action, and one closing sentence naming the specific evidence that would change this read if it came in before the forecast call.Customize
Optional — swap in your own details for the highlighted parts above.
Why this works
The five-category checklist is built from documented, recurring reasons enterprise deals slip — single-threading, verbal-yes-without-process, and unconfirmed next steps are the same signal categories revenue-intelligence platforms like Clari and Gong quantify from call and email metadata at scale, and this prompt applies the same underlying logic manually from what a rep already knows, without needing that instrumentation to reach a similar diagnosis. Requiring one supporting sentence grounded in the actual deal details for each category, rather than a bare applies-or-not judgment, forces the assessment to stay falsifiable against the real facts of this specific deal instead of collapsing into a generic risk checklist that would read identically for any deal handed to it. Forcing one reason to drive the overall risk level, instead of an averaged score across all five categories, closes the most common failure of a risk write-up that nets out to a comfortable, non-actionable "medium" — a deal that's severely single-threaded but otherwise fine on paper is genuinely high-risk, and averaging that severe factor against four calmer ones produces a number that actively hides the one thing a manager reviewing the forecast most needs to see. The explicit instruction to name a single next action rather than a list matters for the same reason a triage room treats one critical patient differently from five stable ones: a list of five equally-weighted to-dos diffuses accountability and attention across all of them, and in practice a rep under time pressure will pick whichever is easiest to do, not whichever most reduces actual risk, so forcing a single answer removes that easy-option escape hatch. The explicit instruction to be willing to downgrade the forecast counters a specific and well-known sunk-cost dynamic in forecast calls: a rep who has already mentally or verbally committed a deal to their manager has a structural incentive to round every ambiguous signal upward rather than down, and an assessment tool that doesn't explicitly counteract that incentive will tend to confirm whatever category the rep walked in wanting to hear.
What you get back
Overall risk: HIGH. Driving reason: the economic buyer has never joined a call and everything about their position is secondhand from the ops manager — this is a single-threaded deal wearing a "commit" label. Single next action: get a call booked with the named economic buyer before this stays on the forecast at this stage.
Verified against
ChatGPT GPT-5.1 · 2026-08-03
Claude Sonnet 4.5 · 2026-08-04
Changelog
- 2026-08-04 — Initial publish, verified against ChatGPT (GPT-5.1) and Claude (Sonnet 4.5).
Need this built into your business?
If a prompt isn't enough — Google Ads management, built and maintained for you — that's Scult's day job.
EXPLORE GOOGLE ADS MANAGEMENT
