War-game a renewal forecast model before stages, confidence bands, and commit rules harden into how every renewal gets booked.
Renewal forecast models fail when the stages invent confidence finance never measured, when commits dual-count the same account as retained and as at-risk, when CS still promises save rates in private Slack, and when RevOps cannot show which stages actually cleared under the published definitions. A neat forecast waterfall is not evidence.
Freeze one model
State the proposed stage definitions and confidence bands, which segments and products it covers, who owns stage hygiene and override audit, what success looks like after two quarters of forecast accuracy and late-stage flip rate, and the kill criteria if shadow commits or stage disputes past a named threshold. Attach the current stage distribution, sample override logs, retention samples, and the measured path from early signal to booked renewal. If CS, finance, and sales disagree on whether the model buys foresight or just deferred blame fights, reconcile before seating.
Name the decision you will make if the war game finds nothing new, and the hold criteria if any stage still lacks a named owner or a verified override path.
Hostile seats
- CFO. Fear that "commit" still means a quiet hope mid-quarter.
- CS lead. Health math that looks clean while the same ten accounts carry every save.
- Sales. Talk tracks that invent expansion that the renewal stage never wrote.
- RevOps. Stage macros that still host retired confidence language.
- Skeptic. The claim that looks strongest and is least sourced.
Optional board seat if public retention guidance binds the form. Same pack for every seat.
Private pass
Feed Pingpong the model draft and exhibits. First pass steelmans the stage ladder. Later passes attack from the seats above. Final pass turns surviving objections into clearer definitions, a published override SLA, or a hold. Delete invented "renewals already behave this way" claims and dual-counted accuracy rates.
Force a day-after narrative: what happens if a large account screenshots mismatched renewal terms, if CS still teaches unwritten save paths, or if finance cannot list the open at-risk cohort. If those stories are stronger than your mitigation plan, fix the package before you publish it. Separate a renewal forecast model from a renewal playbook or a budget reforecast so each decision stays testable.
Ask every seat to mark which overrides are optional in practice. Optional stage audits become skipped within a month if nobody names the owner and the time box. Related: pretend you are the customer success lead, pretend you are the RevOps lead, pretend you are the CFO, and the war-game decisions hub. Process: how to run a Pingpong.