How do you coach reps to commit deals they can actually close?

The Tuesday that exposes everything

Picture a manager with six AEs and a quarter that ends in five weeks. On Tuesday morning the roll-up says $1.4M in Commit. The manager's own number is $980K. Everyone is calm. Four weeks later, $610K closes, and the post-mortem produces the same four sentences it produces every quarter: "procurement took longer than they said," "our champion left," "they pushed to next quarter for budget reasons," "the CFO got involved late."

None of those are surprises. Every one of them was knowable in week one. The champion who left was a champion nobody had backed up with a second thread. The CFO who "got involved late" was the economic buyer the rep had never met — the rep had been selling to a director who kept saying "I'll take it upstairs." Procurement took eleven weeks because nobody had asked how long procurement takes at a company that size. The budget push happened because there was no compelling event, so nothing forced a decision in that particular quarter rather than the next one.

Here is what makes this a coaching problem and not a forecasting problem. If you sit down with that manager and walk the six inflated deals, you will find that in five of six cases the rep could not answer a basic factual question about the deal. Not "was the rep optimistic" — could the rep name the person who signs? Could the rep tell you what happens to the buyer's business if they do nothing for six months? Could the rep produce a calendar invite for the next conversation? The answer is usually no, and the rep did not experience that as a gap, because nobody ever told them that Commit requires those answers.

So the real failure is upstream of the rep. Somebody built a CRM with a picklist that had the word "Commit" in it and never wrote down what earns it. The rep filled it in with feeling because feeling was the only available input. Then quarter-end arrived, the number missed, and the organization concluded it had an optimism problem when it had a definition problem.

The frame that fixes this is simple: commit is a claim, and claims require evidence. A rep who moves a deal to Commit is making a public assertion — "I would bet my number on this closing in this period." Your job as a manager is not to argue with the assertion. Your job is to ask the rep to defend it, in front of you, against criteria you both agreed on before the quarter started. When the evidence holds, you protect the deal and go clear blockers. When it does not, the deal moves back to Best Case and the conversation converts into a short list of things the rep goes and gets by Friday.

That reframe changes who is doing the work. In the bad version, the manager is the skeptic and the rep is the defender, and every forecast call is an argument between two opinions. In the good version, the checklist is the skeptic. You are on the rep's side of the table, helping them find the missing piece before it costs them the deal. Reps stop hiding risk because surfacing risk no longer means losing an argument with you.

One more thing about the Tuesday scenario. The manager in that story had access to call recordings for every one of the six deals and did not open a single one. The rep's narrative was the only input to a $1.4M forecast. This is the cheapest fix available: when a rep says "the CFO is on board," you can go listen to whether the CFO ever spoke. That is not surveillance, and framing it that way to your team is a mistake — it is the same thing a good editor does when a reporter says a source confirmed something. You check the tape.

How the mechanism actually works

Before you correct a rep, diagnose why they over-committed. There are four root causes and they need four different responses, and the single most common mistake in sales management is applying the skill-gap fix to a system problem.

Knowledge gap. The rep does not know what your stage definitions mean. They labeled it Commit because the deal felt further along than the Pipeline deals. Fix: teach the definitions, and put them somewhere the rep sees them — help text on the CRM field, a pinned doc, the first slide of every forecast call. This is a twenty-minute fix that managers skip because they assume the definitions are obvious. They are not obvious. Write them down.

Skill gap. The rep knows what Commit requires and genuinely cannot get there — they do not know how to ask for access to an economic buyer, how to build a mutual close plan, how to secure a real next step instead of "I'll follow up next week." Fix: role-play, call review, and shadowing. This is the only one of the four that responds to practice.

Will problem. The rep is padding Commit to look busy, or hiding a thin pipeline behind a fat forecast, or avoiding a hard conversation with you about a quarter that is not going to work. Fix: this is a directness conversation, not a training conversation. It goes badly if you have ever punished honesty, which is why the pitfalls section below matters so much.

System problem. Your stages have no exit criteria. Your comp plan or your internal culture rewards a fat Commit. Your forecast call is a public performance where the rep with the biggest number gets praised. In that environment inflating Commit is a rational response to the incentives you built, and no amount of one-on-one coaching will out-run the system. Fix the system.

The diagnostic runs fast once you know the branches. Ask two questions — can the rep name the economic buyer and the decision date, and is there a confirmed mutual next step on both calendars — and you have already separated most cases.

Now the conversation itself. Run it inside the weekly 1:1 and structure it as Goal, Reality, Options, Will — the GROW model. The reason to use a structure at all is that it stops you from delivering a verdict. You never say "this deal is bad." You make the rep walk it against the criteria and let the gaps surface on their own, because a gap the rep finds is a gap the rep remembers.

Goal — set the frame in one sentence. "When you put something in Commit you're telling me you'd bet your number on it. Let's pressure-test whether this one earns that. I'm not trying to knock it out — I want it to survive."

Reality — the checklist questions, asked verbatim, answers written into the CRM as you go. Who is the economic buyer and have you personally spoken with them? What is the compelling event that forces a decision by a specific date, and what happens to them if they do nothing? Where did the close date come from — them or you? What is the paper process, including legal, procurement, and security review, and how long does each step take at a company this size? What is our confirmed next step and is it on both calendars right now? Who else is on the buying committee, and who could say no after you think you've won? What do the call recordings show the buyer said about budget and timing, in their words?

The scoring rule is what gives this teeth. Any answer containing "I think," "they seemed," "I'm pretty sure," or "they're excited" fails that criterion. Not because the rep is lying — because a commit is built from names, dates, and quotes, and those four phrases are the linguistic signature of an assumption. Reps learn this fast. Within two or three cycles they stop saying "I think" out loud and start noticing when they are about to.

Options — convert the gap into action. "So we're missing a confirmed economic buyer and a real next step. That reads as Best Case, not Commit, and that's fine. Give me two ways to get to the CFO before Friday. Walk me through the multi-thread plan."

Will — lock the specific commitment and the check-in. "By Friday you'll have a calendar-confirmed call with the CFO and a draft mutual close plan in their inbox. I'll check Thursday. If both land, this graduates to Commit and I'll back it."

Notice you never overruled anyone. The MEDDIC elements — metrics, economic buyer, decision criteria, decision process, identified pain, champion — did the work, and the rep re-rated their own deal. That is the durable version. A deal you demote by authority gets re-inflated the moment you are not looking.

Around that weekly conversation, run a fixed cadence so it does not depend on your memory. Weekly: 25 minutes per rep, top three Commit deals against the checklist, re-rated live in the CRM during the call, not after. Bi-weekly: one closed deal reviewed win/loss — did our Commit signals predict the outcome, and if not, which criterion was wrong? The checklist is a hypothesis about what predicts closing, and it should get tuned. Monthly: a per-rep forecast-accuracy scorecard, reviewed calmly as data, never as a scolding. For a new AE, a 30/60/90: days 1–30 they shadow your deal coaching and learn the definitions, days 31–60 they present and you challenge, days 61–90 they run their own commit logic and you spot-check via call data.

Real numbers, ranges, and benchmarks

Lagging quota attainment tells you nothing in time to change anything. By the time a rep misses, the coachable moment was eleven weeks ago. Track leading indicators instead, and pick thresholds you are willing to act on.

Commit accuracy. Of the deals sitting in Commit at the start of the quarter, what percentage actually closed inside it? This is the master metric. A healthy, mature rep on a transactional cycle should land in the high eighties or better; a rep working six-month enterprise cycles with real procurement exposure will run lower, and that is not automatically a coaching failure — it is a longer chain of things that can move. What matters is that the number is stable and known. A rep at a consistent 75% is forecastable; you multiply and move on. A rep who swings between 95% and 50% quarter to quarter is the actual problem, because nothing you do with their number is reliable.

Slip rate. Of Commit deals, what share pushed to the next quarter rather than closing or dying? Slip is the cleanest early signal of inflation, and it has a distinctive shape: slip clusters in the last two weeks of the quarter, because that is when the deals that were never going to close in-period finally admit it. If your slip is back-loaded, your commits were dated by hope rather than by a compelling event.

Next-step coverage. What percentage of Commit deals have a calendar-confirmed mutual next step with a date, an owner, and an accepted invite? In practice this single metric predicts more than any other, and it is the easiest to audit — you can check it in a CRM report in about ninety seconds without talking to anyone. A Commit deal with no next step on the calendar is not a deal, it is a hope with an amount field. Set the standard at 100% and enforce it mechanically.

Economic-buyer access. What percentage of Commit deals include a logged interaction with the person who actually signs? Reps consistently overestimate this, which is exactly why you check it against calendar and call data rather than asking. The gap between "I'm working with the decision maker" and "I have met the person whose signature is required" is where most late-stage surprises live.

Stage-to-stage conversion. Does Commit convert to Closed-Won at a stable rate? Stability matters more than height. A stable rate is a forecast input; an unstable one means the label is noise and you cannot build anything on top of it.

Forecast call variance. The gap between what the rep called and what landed, trended across six or eight quarters. Shrinking variance is the proof your coaching is working, and it is the only metric on this list that measures you rather than them.

Two practical notes on instrumenting this. First, sample sizes are small — a rep closing eight to fifteen deals a quarter gives you a noisy accuracy number, so read the trend across three quarters rather than reacting to one. One bad quarter is weather. Three is climate. Second, resist the urge to build a rich scorecard. Four metrics that get reviewed every month beat twelve that get admired once and then ignored. If you are choosing, take next-step coverage and commit accuracy and start there.

There is one number worth computing that most teams skip: the average age of a deal in Commit. If deals sit in Commit for eight or ten weeks in a quarter-length cycle, reps are promoting them far too early — Commit has become a synonym for "deal I like" rather than a statement about timing. Deals should enter Commit late and leave fast. When you see the average age climbing, the fix is usually a stage-definition problem, not a rep problem.

And there is one number to compute about yourself. Track how many of your weekly deal-coaching sessions actually happened versus were scheduled. Managers who run this well are usually at ninety percent or better. Managers whose forecast accuracy is bad are usually in the fifties, because deal coaching is the first thing that gets bumped when the quarter gets loud — which is precisely when it matters most.

Trade-offs, alternatives, and the adjacent systems

A commit checklist is not free, and pretending otherwise is how these programs die in month three. Here are the real trade-offs and the alternatives worth considering.

Rigor versus speed. A seven-criterion checklist run against every Commit deal is thorough and expensive. On a transactional motion with dozens of small deals per rep per quarter, that is unworkable — the cure costs more than the disease. Scale the instrument to the motion: for high-velocity SMB, two or three criteria applied only to deals above a dollar threshold; for enterprise, the full checklist on every deal, because a single deal can carry a quarter. The wrong move is running enterprise ceremony on transactional volume and then abandoning the whole thing when reps revolt.

Checklist versus scoring model. Some teams prefer a weighted score to a set of yes/no gates. Scores are more nuanced and much easier to game, because a rep who wants a 70 can find the two soft criteria that get them there. Binary gates are blunt and honest. Start binary. If you graduate to scoring later, keep at least the economic-buyer and next-step criteria as hard gates that no amount of score can override.

Manual inspection versus automated flagging. Conversation-intelligence and revenue-intelligence tooling can surface discrepancies between the rep's story and the recorded reality, and can flag deals with no recent buyer-side engagement. That is genuinely useful and it is also where teams over-rotate. The tool tells you which deals to look at; it does not tell you what to say. Automated risk scores that nobody discusses in a 1:1 change no behavior at all — they just produce a dashboard that everyone learns to ignore.

Manager-enforced versus peer-enforced. A pure manager standard runs on your energy and collapses when you are on PTO or the quarter gets loud. Peer review — each rep presents one Commit deal in the team meeting and the room votes Commit, Best Case, or Pipeline against the criteria — scales better and lands harder, because reps discount a manager's skepticism and do not discount a peer's. The trade-off is real: peer review only works on a team with enough trust that being challenged is not humiliating. On a low-trust or heavily-stacked-ranked team, do not attempt it; it becomes a status contest and reps start protecting each other.

Coaching versus consequence. Some teams add escalating friction for repeated bad commits: first, a short written "what I missed" shared with the team; second, manager approval required to move anything to Commit for a week; third, a full evidence walkthrough before any commit for a month. This works when it is framed as guardrails and lands as punishment when the culture is already fearful. If a rep is afraid of you, adding consequences guarantees they hide deals rather than qualify them. Fix the fear first or skip this entirely.

Three adjacent systems determine whether any of this survives contact with reality, and they sit outside the coaching conversation.

Stage definitions in the CRM. If the exit criteria live only in your head or in a slide deck, they are not real. Put them in field help text, in validation rules where you can tolerate the friction, and in the required-fields list for the Commit category. RevOps owns this, and it is the highest-leverage thing RevOps can do for forecast accuracy — far higher leverage than another dashboard. A required "economic buyer contact" lookup on any opportunity in Commit does more work than six months of exhortation.

Compensation and culture. If your quarterly kickoff celebrates the biggest Commit number and your forecast call is a public performance, you have priced honesty out of the market. Reps respond to what gets applauded. The counter-move is cheap and visible: publicly praise the rep who demotes a deal in week two, by name, in the team meeting. One instance of that does more than a policy document.

Hiring and performance management. Occasionally the honest read is that this is not a coaching problem. A rep who has been walked through the criteria for two full quarters, has had the calls reviewed, has had the role-plays, and still cannot produce an economic buyer on a single deal is telling you something about fit. Deal coaching is not a substitute for a performance conversation, and stretching it into one wastes both people's time.

Worth naming: the same evidence discipline is what makes cross-functional forecasting work. When marketing plans spend, when finance plans hiring, when customer success plans onboarding capacity, they are all consuming the same Commit number. A forecast built from evidence rather than sentiment is not just a sales artifact — it is the input the rest of the company plans against, and its unreliability is felt three departments away long before anyone traces it back to a picklist with no definition.

Common pitfalls and how to avoid them

Rescuing the deal instead of coaching the rep. The manager who was a great AE jumps on the call, works the CFO, saves the quarter, and teaches the rep nothing except that deals get saved. Next quarter the same rep brings the same gap. If you must join a call, agree in advance on exactly what you will do and what the rep will do, and debrief afterward on the moves you made and why. Otherwise you are running the deal, not coaching it.

Coaching the deal instead of the pattern. Fixing one deal a week without ever naming the recurring gap — "you have never met an economic buyer before week eight on any deal this year" — means the rep repeats it forever. Deal coaching is the vehicle; skill development is the destination. Keep a short running note per rep of the gap that shows up repeatedly, and make that gap the subject rather than the deal in front of you.

No follow-through. You agree on a next step in the 1:1 and never check it. This is the fastest way to teach a team that your standard is decorative. The check does not have to be heavy — a two-line message on Thursday is enough. But it has to happen every time for the first month, or the standard evaporates.

One standard applied to everyone identically. A ramping AE in month two and a senior rep in year four need different conversations. The criteria stay the same; the coaching depth does not. Spending forty minutes walking a veteran through basics is condescending and spending ten with a ramping rep is negligent.

Punishing honesty. A rep moves a deal out of Commit in week three, and you react with visible frustration. You have just taught the entire team — because they will hear about it within the hour — that the safe play is to hold the deal in Commit and let it die quietly at quarter-end where the blame is diffuse. This is the single most destructive pattern on the list, and it is almost always unintentional. Watch your face on the forecast call.

Confusing the checklist with the point. The criteria are a proxy for whether the rep understands the deal. A rep who has technically satisfied all seven items and still cannot explain in plain language why this buyer will spend money this quarter has not earned Commit — they have filled in fields. Ask the plain-language question last, every time: "Tell me why they buy now instead of in six months." If the answer is not crisp, the fields were theater.

Letting the forecast call double as the coaching call. A roll-up review with the whole team present is the wrong venue for developmental feedback. Reps perform rather than think, nobody admits doubt, and the manager gets the most confident version of every story. Keep the number conversation and the skill conversation in separate meetings, with separate tones.

Never revising the criteria. If you run win/loss reviews and discover that a criterion has not predicted a single outcome in two quarters, delete it. Checklists that never change become bureaucracy, and reps can tell the difference between a tool that is maintained and a form that is filed.

Treating call recordings as a gotcha. Pulling a transcript to catch a rep in a contradiction poisons the well permanently. Use recordings the way a coach uses game film — watched together, focused on a specific moment, aimed at a specific skill. "Let's listen to the last four minutes where you proposed the timeline" is coaching. "I listened to your call and you lied to me" ends the relationship, and the rep will simply stop recording calls.

Skipping the diagnosis. Almost every pitfall above traces back to the same origin: the manager decided the rep was optimistic and applied the optimism fix, when the actual cause was an undefined stage, a comp incentive, or a genuine skill gap that practice would have solved. Spend the first ninety seconds of every over-commit conversation figuring out which of the four causes you are looking at. It changes everything you do next.

Related questions

How is Commit different from Best Case?

Commit means you would bet your number on it — evidence-backed, closing this period. Best Case means it could close if specific known things break your way. The distinction is not confidence level; it is whether the required facts exist yet. Missing evidence means Best Case, regardless of how good it feels.

Should reps set their own commit or should the manager?

Reps set it; managers audit it against the criteria. A manager-set commit teaches nothing and gets quietly re-inflated. The rep re-rating their own deal after failing the checklist is the entire mechanism — ownership of the judgment is what transfers the skill.

How long before commit accuracy improves?

Expect one full sales cycle before behavior changes and two to three quarters before the metric moves meaningfully. The leading indicators — next-step coverage, economic-buyer access — respond within weeks and are what you should watch early. Judging the program on one quarter's accuracy number will mislead you.

What if a rep's pipeline is too thin to commit anything honestly?

Then the honest forecast is a low one, and that is a pipeline-generation conversation, not a commit conversation. Letting a rep inflate Commit to paper over thin pipeline delays the real problem by a quarter and makes it worse. Address coverage directly.

Does this work for renewals and expansions?

Yes, with different criteria. Renewals hinge on product usage, executive sponsor continuity, and budget-cycle timing rather than compelling events. Keep the structure — written criteria, evidence-backed, reviewed weekly — and swap the specific questions for ones that fit the motion.

FAQ

What exactly is a commit checklist?

A short set of yes/no criteria a deal must satisfy before it can be labeled Commit. Typical items: a named economic buyer the rep has personally spoken with, a compelling event with a date, a documented paper process, a calendar-confirmed mutual next step, and a close date the buyer supplied rather than the rep. Each must be provable with a name, a date, or a quote — not a gut read. Keep it short enough that a rep can recite it from memory.

How do I handle a rep who keeps committing deals that fall through?

Diagnose first: knowledge gap, skill gap, will problem, or system problem. Then run the weekly 1:1 against the checklist, moving deals back to Best Case when evidence is missing and assigning specific actions with dates. Track the pattern across deals rather than fixing them one at a time — if the same criterion fails every week, that is the coaching subject.

Can call analysis really help catch false commits?

It helps considerably. When a rep says the CFO agreed to the budget and no such conversation appears in any recording, you have a concrete coaching moment grounded in observable behavior rather than opinion. Use it as game film reviewed together, not as evidence in a prosecution — the moment it feels like surveillance, reps stop recording and you lose the input entirely.

How often should I review deals with each rep?

Weekly, roughly 25 minutes, focused on the top few Commit deals. Earlier-stage pipeline needs a lighter monthly pass. The weekly cadence matters more than the duration — a consistent short session beats an occasional long one, because the goal is a habit of self-qualification rather than a periodic audit.

What if a rep insists their gut feeling is enough?

Acknowledge that experienced instinct is real and then point out that it is unauditable — you cannot coach it, hand it off, or plan a business on it. The checklist does not replace judgment; it makes judgment inspectable. Reps with genuinely good instincts usually clear the criteria easily, which is a useful thing to point out to them.

How do I get the team to adopt this without it feeling like bureaucracy?

Let them help write the criteria, keep it to five items or fewer, and show one concrete example of a deal that felt great and died on a gap the checklist would have caught. Then hold yourself to the same standard publicly — demote your own optimistic roll-up number in front of the team once, and adoption stops being a fight.

Sources

• MEDDIC Academy — What is MEDDIC
• Harvard Business Review — The Best Sales Managers Don't Chase Revenue
• RAIN Group — Sales Coaching Tips
• Gong — Sales Forecasting
• Clari — Sales Forecasting
• Salesforce — What Is Sales Forecasting
• Winning by Design — Resources
• HubSpot — Sales Forecasting Guide

Related on PULSE

• [How do you give sales reps feedback they'll actually act on?](/knowledge/cg0005)
• [What is the most effective question to determine if a rep is relying too heavily on discounts to close deals?](/knowledge/cg0876)
• [How do you validate that the person you are talking to actually has decision-making authority?](/knowledge/cg0953)
• [How do you build a call-review scorecard that managers actually calibrate on?](/knowledge/cg0931)
• [How do you run a weekly 1:1 that actually improves rep performance?](/knowledge/cg0804)