How do you run a deal-coaching session that actually moves the deal?

The outcome you should expect from a session that actually moves the deal

A deal-coaching session has two outputs, and if you only get one of them you did not coach — you forecast. The first output is deal movement: something concrete changes in the opportunity within seven days. A new stakeholder is engaged. A number lands in the business case where there was an adjective. A next meeting gets a date and an invite rather than a "we'll circle back." The second output is skill transfer: the rep can run that same play unprompted on a different deal next month without you in the room.

Managers who chase only the first output become deal rescuers. They spot the missing economic buyer, they draft the email, they join the call and carry it, and the deal closes. Everyone celebrates. Then the same rep brings the same structural gap on a new logo the following quarter, and the manager rescues again, and the manager's calendar fills with rescues until there is no time left to coach anyone. The rescue habit is the single most common failure mode in first-line sales management, and it feels productive the entire time it is happening.

Managers who chase only the second output become philosophers. The session is a lovely conversation about discovery technique, the rep nods, nobody writes anything down, and the deal sits exactly where it sat. Skill without a committed action is a workshop, not coaching.

So define the outcome precisely before you walk in: one named gap, one rehearsed play, one dated commitment, one agreed proof point. That is the deliverable. Everything else in the thirty minutes is in service of it. If you end a session and cannot state those four things in a sentence, the session did not happen — you had a status call with better lighting.

There is a downstream effect worth naming because it is where RevOps and frontline coaching meet. Sessions that produce dated commitments generate clean data. When a rep leaves a session with "I will get the CFO on a 20-minute call by Thursday and I will log it as a meeting with the economic buyer," your CRM now contains an event that means something. Multiply that across a team and forecast accuracy improves not because anyone got better at forecasting, but because the underlying activity records became honest. Coaching quality is upstream of pipeline data quality, which is upstream of forecast credibility. Most revenue leaders try to fix that chain from the wrong end — they buy a forecasting tool to compensate for coaching that never produced a real next step.

The parallel outside sales is instructive. A good code review does not just fix the bug in the pull request; it leaves the author able to spot the class of bug themselves. A good clinical supervision session does not just diagnose the patient; it sharpens the clinician's diagnostic reflex. Same shape: fix the case, upgrade the practitioner. In every one of those disciplines, the senior person who does the work themselves is considered to be failing at the supervision part of their job, no matter how well the individual case turns out. Sales is the only one of these fields where taking the work over is routinely praised.

What drives the outcome: diagnosing skill, will, knowledge, or a dead deal

Before you coach anything, separate the deal problem from the rep problem. They look identical in a pipeline report and require completely different interventions.

A stalled deal with a clear, nameable cause — nobody has engaged the person who signs, there is no metric attached to the business case, the last three meetings ended without a date — is a skill or knowledge gap. That is coachable in a session. You can teach a multithreading play. You can build a compelling event. You can rewrite a soft close-of-call ask into a firm one and have the rep say it out loud until it sounds natural.

A deal the rep refuses to qualify out quarter after quarter, despite knowing exactly what is missing, is usually a will problem. More technique will not fix it. The rep can already name the gap; they are avoiding the conversation that would expose it. Coaching a will gap looks different — it is about courage, about the emotional cost of hearing "no," about the rep's fear that killing the deal leaves them with an empty pipeline and an uncomfortable one-on-one. Sometimes the honest, kindest coaching move is teaching the rep to disqualify fast and reinvest those hours in something real.

And sometimes there is no gap at all. The deal is dead, the rep knows it, and the only thing being coached is denial.

The diagnostic sequence below runs in about four minutes at the top of a session and prevents the most expensive mistake in coaching: spending thirty minutes teaching technique to someone whose problem was never technique.

The mechanism that makes this work is that it forces the rep to produce evidence rather than narrative. "It's going well, they're really interested" is a story. "The VP of Ops told me on Tuesday that churn is running at fifteen percent and that number goes to the CFO monthly" is evidence. The four-question version of this diagnostic is worth memorizing because it converts stories into evidence in under five minutes.

What exactly has changed since our last session? This forces a concrete event — a meeting outcome, a document shared, a stakeholder added. "We had a good call" is not a change. "They sent me their current vendor's renewal date" is.

Who in the buying committee has committed to a next step, and who hasn't? This exposes whether the rep is talking to a champion who cannot buy. Champions are wonderful and they are not signatures.

What is the one metric the economic buyer uses to justify this spend? If the answer is "efficiency" or "better visibility," there is no business case. If the answer is a number with a unit attached, there is one.

If you had to close this by quarter-end, what's the single biggest blocker you'd remove first? This forces prioritization. A rep who lists three blockers has not isolated the critical path, and a rep who cannot isolate the critical path will spread effort evenly across all three and move none of them.

Write the rep's answers down verbatim. Verbatim matters — the hedges and the qualifiers are the data. When a rep says "I *think* Sarah is probably the one who signs off, at least that's what David implied," you have just learned that nobody has confirmed the decision process, and you learned it from the rep's own sentence rather than from an accusation.

The silence technique carries more weight here than any question. When the rep gives a soft answer, do not fill it in. Wait. Then: "Say more — what makes you confident about that?" The gap surfaces on its own, in the rep's voice, which means the rep owns it. Risk you dictate gets defended. Risk the rep discovers gets fixed.

Session structure, the weekly loop, and the drills that make it stick

Thirty minutes. Two or three priority deals per rep per week, not the whole pipeline. Depth beats breadth every time, and the manager who tries to review every open opportunity has recreated the forecast call they were trying to escape.

The interior of those thirty minutes: five minutes for the rep's own read, uninterrupted. Ten minutes pressure-testing the single biggest gap against a real qualification frame — MEDDIC, Command of the Message, whichever your team actually uses, but one of them, consistently. Ten minutes rehearsing the exact words. Five minutes writing the committed action, the owner, the date, and the proof.

Pull the actual call recording before the session. Gong, Chorus, and Clari all surface this, and the point is not the tooling — the point is that you coach what happened rather than what the rep remembers happening. Memory is reconstructive and reps reconstruct favorably, not dishonestly but inevitably. A recording ends the debate about what was said in eleven seconds.

The GROW model gives the conversation its spine, anchored to the deal rather than floating in the abstract. Goal: "Walk me through in one sentence why this deal closes and when." Reality: the MEDDIC questions above, with the rep grading their own deal before you grade it. Options: "Given that you've never met the economic buyer, what are two ways you could earn that meeting this week?" — the rep generates the plays, and only after they have proposed do you add one of your own. Will: "So what's the one next action, who owns it, by when, and what will we both look at next Tuesday to know it worked?"

Then rehearse. This is the step that gets cut when the session runs long, and cutting it is why next week's follow-up so often produces "I didn't get to it" or "it didn't come up naturally." The rep knew what to do. They had never said the words out loud, so under real pressure the words did not come.

The seven-minute rehearsal protocol: pick the one hard conversation the rep must have before the next check-in. The rep plays themselves, you play the customer, three minutes, no notes, no pausing to think. Then feedback on exactly two things — did they ask for the specific next action, and did they handle the likely objection without backing down. Yes or no on those two points only, no general praise. Run it again for two minutes with the feedback incorporated. If they still hesitate, run it a third time. The goal is not a polished performance; it is that the rep can deliver the core ask without stumbling.

Beyond the rehearsal, a small rotation of drills keeps skills from decaying between sessions. The sixty-second deal pitch: the rep has one minute to convince you the deal is real using MEDDIC, and if they cannot name the metric and the economic buyer in sixty seconds, that is the week's drill. Call-review scorecards built on three behaviors only — talk ratio, the value question, the close-of-call next step. Three, not thirty; a thirty-item scorecard is a document nobody uses twice. Objection role-play with swapped seats, where you play the skeptical CFO on "we don't have budget this year," and then you swap so the rep plays the CFO and has to feel the buyer's logic from the inside. Disqualify drills once a month, where each rep names one deal they should kill and rehearses the script to do it gracefully.

That last one deserves emphasis because it is the drill managers skip. Building the will to protect pipeline integrity is harder than building the skill to advance a deal, and it pays back in forecast accuracy immediately.

Benchmarks, realistic ranges, and what to measure

Quota is a lagging indicator. It tells you nothing about whether last week's session worked, because the deal you coached in March closes in June and by then a dozen other variables have moved. Coach the leading indicators of behavior change instead.

Multithreading depth — average contacts genuinely engaged per open opportunity, where "engaged" means a two-way interaction, not a CC on an email. Track the trend per rep, not against an industry number. If a rep's average was 1.4 contacts in January and 2.6 in April, the access coaching landed. Enterprise deals with large buying committees will run structurally higher than transactional deals; comparing a mid-market AE to an enterprise AE on this metric produces noise, not insight.

Next-step rate — the percentage of customer meetings that end with a confirmed, dated next step. This is the single most useful predictor of whether deals move, and it is almost entirely within the rep's control, which makes it ideal for coaching. It is also trivially auditable: open the calendar. Either the next meeting exists or it does not.

Stage conversion and slip rate, measured against the rep's own baseline rather than a team average. The question is not "is this rep above average" but "is this rep better than they were sixty days ago." Team averages hide the only thing coaching can affect.

Qualification accuracy — the gap between what the rep forecast and what actually closed, narrowing over a quarter. A rep whose commit was fifty percent accurate in Q1 and eighty percent accurate in Q3 has learned to qualify, which is worth more than a rep who simply got luckier.

Coaching adherence — did the rep complete the committed next-best-action by the agreed date? This is the cheapest and most revealing metric on the list. Sustained low follow-through means no coaching is happening regardless of how many sessions appear on the calendar.

Coached disqualifications — deals killed after a coaching session, tracked as a positive. This one is counterintuitive enough that most teams never instrument it, and it is the clearest evidence that coaching is teaching reps to qualify rather than to chase.

On realistic ranges, resist the urge to import benchmarks from a vendor blog and hold your team to them. The honest guidance: measure your own team's current state for four to six weeks before setting any target, because the variance across segments, deal sizes, and sales motions is enormous. A two-week enterprise sales cycle does not exist and a two-week SMB cycle is normal; a next-step rate that would be excellent for outbound cold prospecting would be alarming for a warm inbound motion. Your baseline is your benchmark. The only number that transfers reliably across contexts is the direction of travel.

One structural benchmark does hold up: the sixty-day velocity check. If a deal has been in the pipeline for sixty-plus days with no movement on any of the four diagnostic questions — no new stakeholder, no budget conversation, no measurable business case — it is very likely a dead deal the rep is afraid to kill. Coach the disqualification. Rehearse the exact script: *"Based on our last three conversations, I don't see a clear path to a decision this quarter. I'm going to pause my activity here unless you can introduce me to the person who owns this budget by Friday."* If the deal moves after that push, it was alive and you just accelerated it. If it dies, the rep reclaimed hours that were being spent on a fiction. Either outcome is a win, which is what makes the play safe to run.

RevOps has a real job in this loop, and it is not building another dashboard. It is making sure the five metrics above are computable from the CRM without manual tagging, because any coaching metric that requires a rep to remember to fill in a field will be wrong within three weeks. Next-step rate should derive from calendar events. Multithreading depth should derive from contact-role records and activity logs. If the instrumentation demands discipline the team does not have, instrument something else.

Risks, edge cases, and the ways this goes wrong

Rescuing. Already named, still the number one failure. Taking over closes this deal and teaches nothing. The tell is a manager whose personal win rate on assisted deals is excellent and whose team's unassisted win rate is flat.

Coaching to the deal instead of the skill. You fix one stalled opportunity without ever naming the repeatable gap, so the same miss reappears next quarter on a different logo. The correction is a single sentence at the end of every session: "The pattern here is that you're waiting for the champion to offer access instead of asking for it. That's the thing we're working on, not this deal."

No follow-through. A session with no written action, no owner, and no date is a chat. Write it in the CRM or the shared doc during the session, not after. Anything captured only in a manager's notebook does not exist.

One method for every rep. A new SDR needs modeling — watch me do it, now you do it. A veteran AE needs a sounding board and will resent being modeled at. Applying the same coaching method across experience levels reliably alienates your best people while under-supporting your newest.

Mistaking will for skill. More role-play does not fix a rep who avoids hard conversations or refuses to disqualify. That is a performance discussion, occasionally a plan, and delaying it under the banner of "more coaching" is unkind to everyone including the rep.

Reviewing pipeline instead of coaching deals. "Where's this at?" is forecasting. "What's the one gap and how will you close it?" is coaching. If your session opens with a pipeline report on screen, you have already picked the wrong mode.

Coaching the wrong deal. Managers gravitate to the biggest logo, but the biggest logo is often the one with the most complexity and the least transferable lesson. Coaching a mid-sized deal where the gap is clean and the play is repeatable produces more skill transfer per minute.

Manager capacity collapse. Two or three deals per rep per week across eight reps is roughly sixteen to twenty-four sessions. That is not survivable alongside forecasting, hiring, escalations, and your own pipeline reviews. The realistic answer is a tiered cadence — deep sessions for reps in ramp or in trouble, spot-checks for reps who are self-diagnosing — and this is exactly what the rollout plan below is designed to produce.

Recording culture backfiring. Call recordings make coaching evidence-based, and they also make reps feel surveilled if the only time a recording appears is when something went wrong. Use recordings on winning calls at least as often as losing ones. A team that associates Gong with punishment will start scheduling their real conversations off-platform, and you will have destroyed your own evidence base.

The AI-summary shortcut. Automated call summaries and risk scores are useful for triage and genuinely bad as a substitute for listening. A summary tells you the value question was missed. It does not tell you the rep asked it beautifully and the buyer deflected, which is a completely different coaching conversation. Use the summary to pick which four minutes to listen to; do not let it replace the listening.

A practical rollout plan

Do not roll this out as a program with a name and a kickoff deck. Roll it out as a cadence change that compounds.

Days 0–30 — model it. You sit in on the rep's deal reviews and run the question set yourself while they watch. The rep is not expected to perform; they are expected to observe the structure. End every session by naming what you just did and why: "Notice I didn't tell you the fix — I asked twice and you found it." Set up the recording pull as a habit now, before the harder behaviors land, because the logistics are the easiest part and getting them automatic frees attention later.

Days 30–60 — hand over the wheel. The rep runs their own deal review, presenting the diagnosis rather than the narrative, and you coach the *coaching*: did they find the real gap, or the first gap? Did they isolate one thing or list five? This is the phase where most rollouts quietly revert, because it is slower and less satisfying than doing it yourself, and the manager's instinct to jump in is strongest exactly when the rep is closest to figuring it out.

Days 60–90 — shift to exceptions. You review deals that tripped a risk signal or stalled past average stage time, not everything. Sessions get shorter and sharper. Your calendar reopens. By day ninety a competent rep should be self-diagnosing on routine deals and pulling you in deliberately on the hard ones, which is the actual goal — coaching capacity that scales past the number of hours in your week.

Two operational notes that determine whether this survives contact with a real quarter. First, protect the sessions on the calendar the way you protect customer meetings — the first thing that gets moved when a fire starts is coaching, and a cadence that gets moved twice stops existing. Second, run this on yourself. Have your own manager or a peer sit in on one of your sessions monthly and coach your coaching. Frontline managers get less feedback on their core skill than any other role in the revenue organization, and the same rescue instinct you are trying to break in yourself is invisible from the inside.

The adjacent workflows matter too. Deal coaching that produces dated commitments feeds directly into pipeline inspection, which feeds forecast calls, which feeds capacity planning. When coaching is working, the forecast call gets shorter — because the deal-level questions were already asked and answered on Tuesday, the Friday forecast becomes arithmetic instead of archaeology. That is the clearest signal that the cadence has taken hold, and it is the argument that wins budget and calendar space from a CRO who has heard "we need more coaching" too many times to react to it.

Related questions

How is deal coaching different from pipeline review?

Pipeline review asks "where is this deal and will it close?" — it is a forecasting activity aimed at the number. Deal coaching asks "what is the one gap and how will you close it?" — it is a development activity aimed at the rep. Running them in the same meeting means the forecasting always wins.

Should the rep prepare before a deal-coaching session?

Yes, and keep it to three minutes of prep. The rep arrives with the deal named, their own one-sentence read on why it closes and when, and their own guess at the biggest gap. Prep beyond that becomes a presentation, and presentations invite defending rather than diagnosing.

Can you coach a deal you know nothing about?

Usually better than one you know intimately. Ignorance forces you to ask rather than assert, which is the entire technique. The MEDDIC question set works without product or account context — if the rep cannot make you understand the deal, they cannot make a buying committee understand it either.

What if the rep is more experienced than the manager?

Coach the process, not the content. A veteran AE rarely needs your opinion on the play; they need someone to force the diagnosis, hold them to a dated commitment, and notice patterns across their pipeline that they are too close to see. Ask better questions and stop pretending to know their accounts.

How does RevOps support deal coaching without adding admin work?

By making the coaching metrics derivable from data reps already generate — calendar events, contact roles, activity logs — rather than from new fields. Any metric that requires a rep to remember to tag something will be unreliable within a month. Instrument what the system already knows.

FAQ

How long should a deal-coaching session be?

Thirty minutes per rep covering two or three priority deals. Going deep on one gap beats a shallow sweep of the whole pipeline. If a session routinely runs to an hour, check whether it has drifted into status-checking — the length is usually a symptom of narrating deals rather than diagnosing one.

Should I coach the deal or the rep?

Both, in a specific order: coach the rep's skill *through* the deal. The deal is the live case study and the reason the rep is paying attention; the lasting outcome is a behavior they can repeat. Fix only the deal and you will be back next quarter fixing the same mistake on a new opportunity.

What if the rep insists the deal is fine and I see risk?

Do not argue — ask. "Walk me through who signs and when you last spoke to them." Let the gaps surface in the rep's own words. Self-discovered risk gets fixed because the rep owns it. Risk you dictate gets defended, and you spend the session negotiating instead of coaching.

How do I coach deals on a remote or hybrid team?

Pull the actual call recording and review it together on screen rather than relying on memory. Gong, Chorus, and Clari all support this. Record the rehearsal too — the rep can rewatch their own delivery and self-correct, which is a coaching cycle that costs you nothing.

When is a deal beyond coaching?

When there is no compelling event, no economic-buyer access after genuine attempts, and no champion — and that combination repeats across the rep's pipeline. At that point the useful coaching is teaching disqualification, and if the pattern persists across many deals, the management move is a performance conversation rather than another role-play.

How many deals should I coach per rep each week?

Two or three of the highest-value or highest-risk open opportunities. Depth creates skill transfer; attempting to review everything turns the session back into a forecast call and guarantees that nothing gets rehearsed, which is the step that actually changes behavior.

Sources

• Harvard Business Review — Are You Developing Your Sales Team, or Just Coaching Them?
• Gong Labs — research on sales conversations and deal outcomes
• RAIN Group — Sales Coaching research and best practices
• MEDDIC Academy — What is MEDDIC?
• Winning by Design — sales frameworks and resources
• Clari — revenue operations and deal inspection blog
• Salesforce — sales coaching guidance for managers
• Richardson Sales Performance — sales coaching resources

Related on PULSE

• [How do you run a weekly 1:1 that actually improves rep performance?](/knowledge/cg0804)
• [How do you run a role-play session reps don't hate?](/knowledge/cg0172)
• [How do you run a call-review session that reps don't dread?](/knowledge/cg0127)
• [How do you run a cold-call coaching session with live dials?](/knowledge/cg0036)
• [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)