Most Sales Coaching Is Deal Inspection in Disguise

Put the question to a sales manager, what did you coach this week, and pay attention to the shape of the reply. You will usually get a list: three deals, two close dates, one procurement risk, a forecast that moved. Every item is about a transaction. Not one is about a thing a rep did or said.

That conversation has a name, and the name is pipeline management. It is necessary work and it should be on the calendar. It is simply not coaching, and calling it coaching is how an organisation spends years on development without changing a single selling behaviour. This piece is for a manager or a sales leader who wants to tell the two apart, and then run the second one properly.

One note on sourcing before anything else. A great deal of the coaching statistics circulating online trace back to vendor blogs citing each other, with no sample behind them and figures that contradict between pages. Several have been left out of this article for that reason, and the omissions are flagged where they matter.

Why the Habit Is Rational

The habit is not laziness. It is a rational response to what a manager can actually see. Deal status arrives in a dashboard. Selling behaviour is buried in conversations nobody senior was present for, and digging it out costs hours no plan allocated.

There is a quantified version of that time problem, though it comes with a caveat. The Sales Management Association's write-up by Michelle Vazzana, chief executive and a founding partner at Vantage Point Performance, reports an estimate attributed to Ohio University researcher Adam Rapp that managers spend roughly 32% of their time managing their sales teams and 68% on everything else. Treat it as an estimate: the page states no sample for it. The direction is uncontroversial to anyone who has held the job.

What the same page does carry with a stated sample is the scale of the attainment problem. Per that article, a global study of more than 1,000 sales managers found that 75% of them had fewer than 49% of their salespeople on target, and it puts the opportunity cost against top-performing peers at roughly $3.5 million per manager. A separate survey of 515 sales managers, described on the same page, looked at which coaching practices correlated with quota achievement and found high-performing managers getting 30% more of their sellers to quota.

Note what is missing from that page and should be: a figure it offers comparing the performance impact of methodology training against coaching training carries no stated sample, so it is not reproduced here.

Feedback Is Not Automatically Good

Here is the finding that should unsettle anyone running a coaching programme, and it predates every sales vendor by three decades. Avraham Kluger and Angelo DeNisi published a meta-analysis of feedback interventions in Psychological Bulletin in 1996. Per the abstract hosted on Kluger's university page, the analysis pooled 607 effect sizes across 23,663 observations and found that feedback interventions improved performance on average, at d = .41, while more than a third of them made performance worse.

More than a third. Not neutral: worse. And per the same abstract, the result could not be explained by sampling error or by whether the feedback was positive or negative.

The mechanism they propose is the part that matters for sales. Feedback shifts where a person's attention lands, across three levels: learning the task, being motivated to do the task, and thinking about the self. Effectiveness falls as attention moves up that hierarchy, away from the task and toward the self.

Now read a standard deal review through that lens. "Why is this still in stage three?" points a rep's attention at their own number, their own standing and their own competence. That is the top of the hierarchy, where the evidence says feedback works least well. "At eleven minutes you answered the pricing question before she finished asking it, let us try that differently" points attention at the task. Same manager, same half hour, two very different interventions, and the 1996 meta-analysis is the strongest available reason to think the second one works and the first one can actively backfire.

We are using a 1996 paper deliberately. Nothing published since combines that sample with that generality, and the finding has not been overturned. Treating it as old is a mistake people make because the date looks distant.

Coverage Beats Cadence, and That Is an Uncomfortable Finding

Most coaching programmes are designed around rhythm. Weekly one-to-ones, a monthly skills session, a quarterly review. The implicit theory is that if the cadence is right, coaching happens. There is now dated evidence that the cadence is the wrong variable.

Ambition surveyed 312 sales managers, directors, vice presidents and chief revenue officers for its 2026 state of coaching report, published in September 2026. Per the write-up, coaching frequency turned out to be a weaker predictor of quota attainment than coaching coverage, meaning the share of reps who actually get coached. In organisations where 76 to 100% of reps received weekly structured coaching, 54% reported quota attainment above 71%. Where fewer than 25% of reps received it, only 25% reached the same threshold, and 53% landed in the lowest attainment bucket.

Two honest caveats. Ambition sells a coaching platform, so a finding that favours systematic coverage flatters the product. And this is self-reported survey data from leaders, not observed outcomes. Hold it loosely.

Our position, nevertheless, is that coverage is the right thing to manage and almost nobody manages it. The reasoning is the mechanism the same report describes: coaching naturally gravitates toward the reps who ask for it and the deals that are in trouble. Both of those are self-selecting. The rep who asks is already engaged, and the deal in trouble is a deal review wearing a coaching badge. What gets missed is the quiet middle, which is also where most of your pipeline sits. Per that survey, 64% of organisations coach fewer than half their reps weekly and 31% reach fewer than one in four.

So the metric to put on a wall is not sessions held. It is the share of your reps who had a structured coaching conversation in the last seven days. That number is unflattering the first time anybody calculates it, which is precisely why it is useful.

Suppose coverage turned out to be a proxy for something else, such as managers with lighter spans of control or teams in healthier territories, then chasing coverage would be chasing a symptom. That is a real risk with self-reported survey data and it is testable inside one company by comparing coverage against attainment while holding team size constant. Span of control is worth getting right either way, and we take it apart in session length and coach-to-rep ratio.

Four Kinds of Coaching, Only One of Which Is a Deal Review

The vocabulary problem is doing real damage, and the research literature is clearer than the industry is. The study of 515 managers described by the Sales Management Association separated coaching into distinct types rather than treating it as one activity. Per the write-up, those types are call coaching, opportunity coaching, account coaching and territory coaching, examined alongside the frequency, formality and content of the conversations.

Look at that list and the central claim of this article stops being an opinion. Opportunity coaching, which is the deal review, is one of four. A manager doing only that is covering a quarter of the discipline and reporting it as the whole thing.

• Call coaching works on what the rep said and did. It needs a recording and it is the one that changes behaviour.
• Opportunity coaching works on a specific deal. Useful, finite, and it fixes exactly one deal.
• Account coaching works on a relationship over time, including expansion and the people you have never met.
• Territory coaching works on where the rep spends their hours. It is the least glamorous and often the highest leverage.

Keep the deal review. Give it ten minutes and its own slot, and stop charging it to the coaching budget. The weekly rhythm that holds both is covered in our thirty-minute one-to-one system.

Behaviour Hides Inside the Call

Call data exists because the thing you need to coach is invisible everywhere else. Gong examined 326,000 calls that ran ten minutes or longer between sellers and prospects. Per its talk-to-listen study, refreshed with 2025 data, a typical call splits 60% speaking against 40% listening; sellers took 57% of the airtime on deals that closed and 62% on deals that did not; and question counts came in at 15 to 16 on wins against roughly 20 on losses.

The won-against-lost gap is five points, which the page itself concedes is not a large difference. The genuinely useful finding sits underneath it. Per the same dataset, high performers hold roughly the same talk ratio whether they are winning or losing, while low performers swing by 10%, from 54% on won deals to 64% on lost ones.

Sit with that for a second, because it reframes the whole coaching target. The difference between a strong seller and a weak one is not an ideal ratio. It is consistency under pressure. A struggling rep talks more when the deal is going badly, which is the exact moment talking less would help. You cannot see that in a dashboard, you cannot see it in a forecast, and the rep cannot see it in themselves. It only shows up when you line up two recordings, one from a deal that worked and one from a deal that did not.

That is the whole argument for call review in one observation, and the practical method is in coaching discovery calls from recordings. Dan Morgese, director of content strategy and research at Gong, offers a concrete drill on the same page: a two-second rule, pausing for a beat before responding. It is small enough that a rep can actually do it on Monday.

The 43:57 Ratio You Should Stop Quoting

A quick correction, because this one is everywhere. The famous "golden ratio" of 43% talking to 57% listening is real but it is historical. Gong's current page presents it in the past tense as what the 2016 analysis found, alongside the finding that talking more than 65% of the call was linked to lower conversion.

Two things follow. The current figures in that dataset sit at 60:40 overall and 57% on wins, not 43:57, so a team drilling toward 43% is drilling toward a nine-year-old target. And no sample was ever published for the 2016 pass, which leaves the most-quoted ratio in the discipline with no stated n behind it. Quote the 2025 figures, name the 326,000-call sample, and let the old one go.

The Objection Moment, Measured

One more behaviour that only a recording reveals, and it has a number attached. Gong pulled 67,149 call recordings from a database of over 1M calls, all of them demos run over screen sharing, and analysed what happens immediately after an objection. Per that 2018 study, average reps respond with an uninterrupted monologue stretching 21.45 seconds.

The qualitative findings around it are the coachable part. The same analysis reports that stronger sellers pause longer after an objection than during normal conversation while average reps pause less, that they ask clarifying questions rather than assuming they understood the concern, and that the back-and-forth pace of the conversation holds steady for them while it drops sharply for everyone else. The page is candid that this is correlation rather than causation.

Note the date: 2018. It is used here because no newer analysis with a comparable sample on objection behaviour has been published. The behaviour it describes, pausing and clarifying instead of defending, is not the sort of thing that goes out of date.

REKS: Diagnose the Lever Before You Pull It

If the call is where you look, you still need a way to decide what to work on. Winning by Design publishes a diagnostic worth borrowing. Per its REKS framework page, the letters cover results, effort, knowledge and skills, and the exercise is to establish which of the latter three is producing a shortfall in the first. The page argues for working "one lever at a time" so the rep is not swamped, and for one-to-ones that diagnose rather than inventory activity.

Be clear about what this is. The page carries no study, no sample and no effect size. It is a prescriptive framework rather than evidence, and the full document sits behind a form. Borrowed as a diagnostic habit rather than as proof, it is still the most useful four-way split available.

• Effort gap. The work is not getting done in sufficient volume. Coach the calendar and the plan, not the technique.
• Knowledge gap. Thin grasp of the product, the buyer or the market. That calls for instruction, which is a different activity from coaching.
• Skill gap. They can describe the right move and still fumble it live. Only this one needs a recording and a rehearsal.

Pulling the wrong lever is worse than pulling none. Running skills drills at an effort problem reads to the rep as a manager who has not looked at their calendar, which pushes attention straight to the self, where the 1996 meta-analysis says feedback stops working.

A Four-Week Build

1. Week one, measure coverage. Count how many of your reps had a structured coaching conversation in the last seven days. Write the number down before you fix anything.
2. Week one, choose a single stage transition for each rep. The step from an opening meeting to a follow-up one is usually cleanest, being both frequent and early enough to change something.
3. Week two, listen before you meet. Two recordings from that stage, one that progressed and one that did not. Capture verbatim wording with timestamps instead of general impressions.
4. Week two, let the rep go first. Open with what they would change about the call. Self-diagnosis keeps attention on the task, which is the whole point of everything above.
5. Week three, agree one behaviour. One, written in a single sentence, observable on a recording. Two behaviours is zero behaviours.
6. Week four, close the loop. Check the next recording for the specific thing you agreed. An unclosed loop teaches the rep that coaching is theatre.

Repeat with the quiet middle before you repeat with the rep who asks. For the reinforcement side of this, see evidence-based techniques that actually change behaviour, and if you are considering tooling to widen coverage, what full call coverage can and cannot do.

Objections From the Management Line

"I do not have time to listen to calls." Two recordings per rep per fortnight is roughly an hour. The deal review you are protecting instead fixes one deal and teaches nothing, which is the trade you are actually making.

"My reps will feel watched." Some will. Say plainly that you will put your own calls up for review too, and that the conversation is about a sentence at minute eleven rather than about them. The evidence on where attention lands is the reason that framing is not just manners.

"Deal reviews are what my leadership asks for." Then keep them and keep them short. These are different meetings with different purposes and defending one does not require abandoning the other.

"We bought a conversation intelligence tool and nothing changed." Tools raise coverage of recordings, not coverage of coaching. Somebody still has to pick one behaviour and close the loop on it.

Honest Expectations

The first month feels slower and thinner. There are fewer deals discussed and more silences while somebody plays a clip. Some reps push back, and the pushback is usually a reasonable fear rather than resistance to development.

Nobody can promise you a quota outcome from this, and anybody who does is selling something. Behaviour change shows up in recordings within a few weeks because that is where it lives. Win rates move on the clock of your own sales cycle, so judging them earlier than that will tell you nothing. What it does deliver dependably is sight of which behaviours move the needle in your particular calls, and no dashboard has ever offered that.

One Number to Watch

Pick one rep from the quiet middle, one conversion point, and two recordings. Agree a single behaviour and check it the following week. Then calculate what share of your team had that conversation in the last seven days, and treat that percentage as the health metric for your coaching programme rather than the number of sessions on your calendar. If that session felt unlike the pipeline meeting you ran yesterday, you have located the boundary this whole article is tracing.