A short plan both sides own, built backwards from the buyer's go-live date, with a name and a date on every step and the contract as one step in the middle. Keep it under fifteen steps.
Everyone agreed the deal is happening. Nobody can say what happens next week, or who does it.
Write a short plan that both sides own: the steps between today and the day the buyer gets the result they are paying for, with a name and a date on every step on both sides. The contract is one step in the middle, not the end. Build it backwards from the date that is real to the buyer, the day they want to be live, and put buffers where steps usually run long. If you cannot build a believable plan back from their date, the deal is probably not closing this quarter. Keep it short, have the buyer write their own next step, and repeat the agreed dates at the end of every meeting and to every new person who joins.
Nick Cegelski's work-back plan, dated in reverse from a close on 26 June: economic buyer approval on 8 June, finance approval on 10 June, first redlines on 15 June, second redlines on 16 June, legal-to-legal call on 17 June, contract signed on 18 June, and a buffer from 18 to 26 June. He shares it with the buyer rather than keeping it internal, and his test is the one above: no believable work-back plan, no deal this quarter.
Armand Farrokh's "remind and reveal". Remind the buyer of the agreed next steps at the end of every meeting, in every recap email, and for every new person who joins. Reveal more detail as the deal moves: a loose walk-back plan at the end of discovery, a tighter plan once the champion is won, then a detailed one that includes the vendor-review milestones.
The plans in GTMnow's collection. Cloudability's spreadsheet lists the roles on both sides before the names are known, so it doubles as discovery, then the milestone, dates, duration, outcome, and a column the buyer fills in with feedback. Outreach's plan opens with a value summary, says the plan is non-binding, and runs past signature into implementation. GTMnow puts most plans at eight to twelve lines. Dock's template has five sections: an overview with the kickoff date and success criteria, a progress timeline, evaluation, procurement with legal and security review, and implementation.
Keep it short. Tom Williams, who heads Clari Align, says a four-milestone plan can do a lot and treats each milestone as a small project with its own tasks, deadlines and success criteria. MEDDICC's Go Live Plan makes the product going live the goal instead of the deal closing, so the plan carries on past signature.
The M&A version. A sell-side process letter fixes the bid deadline, what a bid must contain and whom to contact, and bidders who miss its requirements can be dropped. The seller owns the timeline instead of each bidder assuming one.
The vendor numbers, and what to make of them. Outreach reports a 26% higher win rate on deals where the seller worked with the buyer on a plan. trumpet reports a 60% win rate on deals with a plan against a 29% baseline, rising to 84% on plans with six to ten completed steps and 92% on plans with eleven to fifteen, and falling back to the baseline past thirty steps, from "hundreds" of its deal rooms. Both companies sell the software that makes these plans, neither publishes a method, and both are correlations: a deal going well is more likely to get a plan and to have its steps completed. The step-count pattern is still useful, because it points against the vendor's own product: past a point, more plan went with worse results.
Help with the buying goes with better deals. Nicholas Toman, Brent Adamson and Cristina Gomez of CEB reported in Harvard Business Review in 2017 that across more than 600 B2B buyers, suppliers who made buying easy were 62% likelier to win a high-quality sale. It is not a test of plans, and CEB sold the advice built on it, but it is the measured case for what a plan does, which is to take the organising work off the buyer.
Nobody has tested work-back plans in selling. Cegelski, Farrokh and Williams agree on the method and publish no measurement.
Planning backwards makes the date honest. Jessica Wiese, Roger Buehler and Dale Griffin ran four experiments in 2016 in which people laid out the steps to a deadline either forwards or starting from the last step. In the fourth, with a real project due within the month, backward planners expected to finish 2.82 days early and finished 2.64 days early. Forward planners expected 3.91 days early and finished 1.98 days early. Planning backwards made nobody faster; it made the date they gave close to true.
A plan that names when and how changes what people do. Peter Gollwitzer and Paschal Sheeran's review of 94 tests found that plans naming a trigger and an action raised follow-through by a medium-to-large amount, where general intentions did not. David Nickerson and Todd Rogers asked 287,000 voters when, where and how they would vote, which raised turnout by 4.1 points, and did nothing in households where the decision was shared, a warning for a plan handed to one person in a buying group. In NHS clinics, Steve Martin and colleagues found that having patients write down their own appointment details cut missed appointments by 18%, and by 31.7% with a notice about how many people attend.
A name on every step. Steven Karau and Kipling Williams' review of 78 studies found that people put in less effort when their own contribution cannot be traced to them, which is the case for an owner on each line.
Why the buyer's own date holds better than yours. In The Strategy of Conflict, Thomas Schelling explains how a public commitment binds the party who made it: backing down costs them. A go-live date the buyer has already promised to their own people is that kind of commitment. A date you propose is not.
Nobody has randomised deals into plan and no plan, and both win-rate figures come from companies selling plan software. The experiments on backward planning and on naming when and how used individuals planning their own tasks, not a buying group holding to a shared date.
Build the plan backwards from the buyer's own go-live date, put a name and a date on every step on both sides, keep it under fifteen steps, and have the buyer write their own next step. Repeat the dates at every meeting. The plan also works as a diagnosis: the step nobody will put their name against is where the deal will stall.