Exit criteria are the short, observable checklist each opportunity must satisfy before it can advance to the next stage. Done well, they stop reps from dragging weak deals forward and give your forecast a factual spine. This guide shows you how to write criteria that hold up under pressure, enforce them in your CRM, and audit them every quarter so they keep working as your motion matures.
Time: 60 minutes
Before you write a single criterion, write a one-sentence definition of what the buyer has agreed to by the time they enter each stage. Qualified means the buyer confirmed a real problem and a willingness to look at solutions. Discovery means the buyer agreed to open their process to you and introduce the people who care. Proposal means the buyer asked for pricing against a defined scope. If you cannot describe a stage in buyer language, the stage is wrong, and criteria built on top of it will decay. Stages named after seller actions, demo scheduled or proposal sent, invite reps to game them. Stages named after buyer commitments force honest answers because the buyer either did the thing or did not.
The useful range is three to four criteria per stage. Fewer than three and the gate is too loose to filter noise. More than four and reps either lie, bypass, or stop updating the record entirely. Each criterion must be binary, observable, and recorded somewhere a manager can audit without asking the rep. Economic buyer identified by name and title is observable. Buyer seems engaged is not. Confirmed budget range in writing is observable. Budget feels real is not. The discipline of writing binary criteria forces you to pick what actually matters. The second-tier signals you had to cut will show up later, as coaching questions in the weekly review, not as mandatory gates in the CRM.
Replace every soft verb in your draft with an evidence verb. Replace understands with confirmed in writing. Replace is interested with scheduled next meeting. Replace has budget with named budget range on a recorded call. The point is not to turn reps into stenographers. The point is to anchor stage advancement to artifacts a manager can review when a deal slips. If the artifact does not exist, the criterion was not met, and the deal does not move. Evidence language also protects reps from themselves. The optimistic read of a buyer meeting fades fast; the recording, the sent email, and the named contact in the CRM do not. Build criteria that survive the end-of-quarter audit, not just the Monday forecast call.
Criteria written in a wiki page nobody reads do nothing. Each criterion must live inside Strkr as a required field on the opportunity, tied to the stage it gates. Decision maker becomes a required contact lookup. Confirmed budget range becomes a required picklist. Pain confirmed becomes a required text or linked note. Then add a validation rule that blocks the stage advancement if any required field for that stage is blank. Reps will complain the first week. By week three the complaints stop and the data gets honest, because the only way to move a deal forward is to go collect the evidence the stage requires. That friction is the entire point. The CRM has stopped being a passive filing cabinet and started being a quality gate.
Real deals do not always march forward in order. Some fast-track past early stages because an executive sponsor pulled them in. Others need to go back because new buying-committee members surfaced and discovery has to restart. Write both policies before launch. Skipping forward should require manager approval and a logged reason, so the system still forces visibility. Reverting backward should be routine and encouraged, because it reflects honest reassessment. Punish silent skips, where a rep drags a deal from early qualification to late-stage negotiation without ever meeting the gates in between. Those deals are either rare genuine fast-tracks or hidden diligence gaps, and the approval requirement surfaces the difference without accusing anyone.
Publish the criteria inside Strkr, next to the stage picker on the opportunity record, not in a slide deck that lives in a shared drive. The rep needs to see the exact checklist for the next stage at the moment they are deciding whether to advance. Add a short rationale under each criterion that explains why it matters, pulled from your closed-lost analysis. Buyers who did not confirm budget range before Discovery lost to no-decision twice as often. That one line turns a rule into a reason, and reps follow reasons longer than they follow rules. Keep the whole document short enough that a rep can scan it between calls. If it needs a table of contents, it is too long.
A pipeline review that does not reference stage criteria is a status update. Rebuild the agenda around the criteria themselves. For every deal above a dollar threshold or in the top five per rep, the manager asks three questions tied to the current stage: which criteria are satisfied with named evidence, which are not, and what is the rep doing this week to close the gap. Reps who cannot answer are told to come back next week with evidence, not opinions. Managers who hold this line see forecast accuracy improve within two quarters because the forecast stops being a story about rep confidence and starts being a readout of which gates have been passed.
Pull the last ninety days of closed-won and closed-lost deals and test every criterion. For each stage, calculate conversion to closed-won among deals that met every criterion versus deals that advanced with gaps. If the two conversion rates are within five points of each other, the criterion is not predictive and should be rewritten or dropped. If the gap is wide, keep it and defend it. Also look for criteria reps bypass most often. A frequently bypassed criterion is either badly written, poorly enforced, or capturing something buyers no longer care about. Any of the three is a signal to redesign, not to tighten the enforcement further. Treat the criteria the way a product team treats a shipping feature: measured, iterated, and never finished.
A change to exit criteria is a change to how deals get moved and how reps get paid. Treat it that way. Any time you add, remove, or rewrite a criterion, run a short enablement session with the team: here is what changed, here is why, here is what good evidence looks like, here is a worked example from a recent closed-won deal. Record the session and link it from the opportunity help panel so new hires onboard against the same standard. Then give reps two weeks to adapt before enforcement tightens. Rolling out a new criterion as a surprise kills trust in the system faster than any amount of bureaucracy. Rolling it out with context, examples, and a grace window keeps the discipline high and the complaints low.
Three or four is the useful range for most B2B motions. Fewer than three and the gate is too loose to filter noise. More than four and reps start marking criteria complete without the evidence to back it up, which quietly destroys forecast accuracy. If you feel you need five, split the stage instead of lengthening the checklist.
A good criterion is binary, observable, and tied to buyer behavior rather than rep activity. A manager should be able to verify it from the CRM record alone in under a minute. Economic buyer identified by name and title is a good criterion. Buyer feels bought-in is not, because there is no artifact a manager can audit.
Enforce them inside the CRM as required fields with validation rules, not as a separate approval queue. Reps feel one friction point at the moment of stage advancement, not an ongoing bureaucracy. Pair the enforcement with a short rationale under each criterion explaining why it matters, pulled from closed-lost patterns. Rules with reasons are followed. Rules without reasons are routed around.
Audit the criteria quarterly against the last ninety days of closed-won and closed-lost data. Compare win rates for deals that met every criterion versus deals that advanced with gaps. Rewrite or drop any criterion where the predictive gap is under five points. Resist mid-quarter changes unless something is clearly broken, because changing criteria breaks historical reporting.
Allow skipping forward with manager approval and a logged reason, so the system still forces visibility. A deal that jumps from early qualification to late-stage negotiation is either a genuine executive-sponsored fast-track or a rep hiding missing diligence. The approval requirement surfaces the difference without accusing anyone. Reverting backward should be routine and encouraged, because it reflects honest reassessment.
No. Build one set of criteria per motion, not per rep. SMB and enterprise motions have different buyer committees, different cycle lengths, and different evidence standards. New-business and expansion motions often look nothing alike at the same stage. Keep each motion's criteria short and specific. Fragmenting beyond motion level makes cross-rep benchmarking impossible.
Strkr turns your exit criteria into required fields, validation rules, and manager review views so deals only advance on evidence. Write the criteria once, enforce them automatically, and audit them every quarter without stitching tools together.
No sales call, no migration consultant, no four-month implementation. Enter your card, get 14 days of the full Pro tier, cancel any time before day 14 with zero charge. Spin up a workspace, import your CSV, and have something useful before lunch.
