How do you build a weekly sales coaching cadence that sticks?

The outcome you should expect

The measurable outcome of a functioning weekly coaching cadence is not a quota lift in month one. It is behavior change that shows up in call recordings within two to four weeks, conversion-rate movement at one specific funnel stage within six to ten weeks, and quota-attainment movement one to two quarters out. Managers who expect the reverse order — quota first — kill the cadence before it can work, because at week three the numbers look identical and the program feels like a waste of a manager's most expensive hour.

Set the expectation explicitly with your leadership and with the reps. In week one you are establishing the mechanic: the block exists, it is not cancelled, and it opens with last week's commitment. In weeks two through four the observable win is that reps arrive prepared — they have listened to their own call before the meeting and have a point of view about it. That single shift, from the manager narrating the rep's call to the rep narrating their own call, is the first real proof the cadence is taking hold. If by week four your reps still walk in cold and wait to be told what happened on their calls, the cadence is a status meeting wearing a coaching costume.

The second outcome to expect is a narrowing of the coaching surface. A cadence that sticks converges: instead of a new random topic weekly, the team drifts toward two or three recurring skill themes — usually discovery depth, multithreading, and next-step discipline. That convergence is a feature. It means you are diagnosing patterns rather than reacting to whatever call you happened to hear last. It also makes group coaching viable, because when six of eight reps share a discovery-question gap, you can run one team drill instead of eight identical 1:1 conversations.

The third outcome is cadence adherence itself becoming a reportable metric. Track the percentage of scheduled 1:1s that actually happened and the percentage of written commitments that were inspected the following week. Those two numbers describe the manager, not the rep, and they are the leading indicator that everything else depends on. A team running above 90% adherence with above 85% inspection has a cadence. A team at 60% adherence does not have a cadence — it has a calendar invite.

Finally, expect the cadence to change what your pipeline reviews feel like. When skill coaching has its own protected slot, forecast calls stop being the place where a manager tries to smuggle in feedback between deal updates. Forecast gets faster and more honest because nobody is defending their competence while defending their number.

What drives that outcome

Four mechanics carry almost all the weight. Everything else is decoration.

The protected recurring block. The block must be a standing calendar series at the same day and time every week, owned by the manager, treated as a customer meeting. The rule that makes this survive a bad quarter is *never cancel, only shorten*. A 45-minute session compressed to 15 minutes still runs the same loop with a 60-second clip and a one-sentence commitment. A cancelled session teaches the rep the cadence is optional, and that lesson is nearly impossible to unteach. Managers who allow the first cancellation in week three almost always report the cadence dead by week eight.

Evidence over memory. Every session anchors on a specific artifact — a 2-to-3-minute call clip from a conversation-intelligence platform such as Gong, Chorus, or Salesloft, a recorded demo, or a written email thread. Without an artifact, the conversation drifts into hypotheticals ("what would you do if a prospect said…"), and hypothetical coaching rarely changes real behavior. The artifact also removes the argument about what actually happened, which is where a surprising amount of coaching time gets burned.

One behavior, not the forecast. The session coaches a single observable behavior — "ask a quantified-impact question on every discovery call," not "book more meetings." A behavior can be inspected in a recording; an outcome cannot be coached directly. Deal strategy and pipeline status belong in a separate meeting. When the two are merged, the urgent (this quarter's deals) always eats the important (next quarter's skill).

The written, inspected commitment. The rep writes the commitment in their own words, in a shared doc that carries forward, and the next session opens by inspecting it. This is the closing loop that converts a nice conversation into a cadence. Without it you get eight thoughtful discussions and zero compounding.

The loop above has one deliberate property: it always returns to inspection. Managers who drop the inspection step and keep everything else still lose the cadence, because the rep quickly learns that commitments are aspirational. Managers who keep only the inspection step and drop the drills still get partial results, because inspection alone creates accountability. If you have to sacrifice a stage under time pressure, sacrifice the drill, never the inspection.

Note also the three exits from the diagnosis branch that leave the cadence entirely. A persistent will problem with a fairly resourced rep is a performance conversation, not another role-play. A territory, comp-plan, or ICP problem is a RevOps and CRO issue — no volume of weekly coaching fixes a rep sitting on a dead patch. Being honest about those exits protects the cadence's credibility. Reps notice when a manager keeps coaching a problem that everyone knows is structural.

Benchmarks and realistic ranges

Use these as planning ranges, not laws. They come from how the mechanics behave in practice, and your own numbers should replace them within a quarter.

Session length: 30 minutes weekly per rep, 45 if you can afford it. Thirty minutes fits an inspection open (3–5 min), a clip and diagnosis (10–12 min), rep-generated options (8 min), and a commitment (5 min). Under 20 minutes you can still run the loop, but you lose the options phase and the manager starts telling instead of asking. Over 60 minutes weekly, attendance decays because the block is too expensive to protect during a heavy quarter.

Span of control: 6–8 reps for a real weekly cadence. At six reps, a weekly 30-minute cadence costs three hours a week plus prep — roughly four to five hours of manager time all in. At ten reps that becomes eight-plus hours weekly and something breaks, usually the newest or lowest-performing reps who need it most. Above eight reps, the honest move is to run weekly 1:1s for a prioritized subset (typically the middle of the distribution, where coaching has the most leverage), biweekly for the rest, and a weekly group call-review session to cover the gap.

Layered rhythm. Weekly 30-minute skill 1:1; monthly 60-minute deep-dive on one theme with three or four scored calls; quarterly reset of development goals and trendlines. New hires get compressed coaching: days 1–30 knowledge and shadowing, days 31–60 supervised reps with short daily micro-coaching, days 61–90 independent calls with the standard weekly cadence.

Time to visible signal. Behavior appearing in recordings: 2–4 weeks. Stage-conversion movement at the coached stage: 6–10 weeks. Win-rate or attainment movement: one to two quarters. Ramp-time improvement for new hires is usually the earliest clean read, because new reps have no established habits competing with the coaching.

Adherence targets. Aim for 90%+ of scheduled sessions actually held and 85%+ of commitments inspected the following week. Below 75% adherence, treat it as a manager-discipline problem and fix the calendar before touching the content. Track both numbers monthly; they are the only two coaching metrics worth putting in front of leadership early, because everything else lags too far to defend the program.

Clip length. Two to three minutes, not a whole call. A full 45-minute recording eats the session and dilutes attention. Pick the 90–180 seconds around the moment that matters — the objection, the pricing question, the missed next step. Managers who insist on whole-call review run out of time to commit to anything.

Prep cost. Budget 10–15 minutes of manager prep per rep per week to find the clip and review last week's commitment. That prep is the single most-skipped input, and skipping it is what turns a coaching session into an improvised chat. Conversation-intelligence tooling cuts this meaningfully by surfacing candidate moments, but it does not eliminate it — a human still chooses which moment matters for that rep this week.

Coaching surface per session: one behavior. Two is tolerable occasionally; three guarantees none of them land. Reps executing under quota pressure can hold one new deliberate behavior at a time.

Risks, edge cases, and failure modes

The manager is usually the failure point. Before diagnosing a rep, diagnose your own discipline. The most common cause of a dead cadence is a manager who cancelled during a busy stretch and never restarted with the same seriousness. The second most common is a manager who kept the block but stopped preparing, so sessions became generic. Both are fixable, but only if named honestly.

Pipeline creep. The cadence's most reliable killer is deal talk leaking in. It starts innocently — "quick thing on the Acme deal before we start" — and within a month the coaching block is a second forecast call. The countermeasure is structural: put forecast on a different day, and when a deal question surfaces mid-session, write it on a parking-lot line and return to the clip. If deal urgency genuinely can't wait, end the coaching session early and run the deal conversation separately rather than blending them.

Coaching the deal instead of the skill. "Send them the case study" wins one deal. "Here's how to handle social proof whenever it comes up" wins ten. Deal coaching feels productive because it produces an immediate action, which is exactly why it crowds out the slower, more valuable work.

Telling instead of asking. Giving the rep the answer is faster and builds nothing. The discipline is to make the rep narrate first and generate two or three options before you add yours — and when you add yours, frame it as one more option, not the verdict. Managers who were strong individual contributors fail here most often, because they genuinely do know the better move and it costs them something to withhold it.

Uniform coaching across a non-uniform team. An A-player needs stretch goals and autonomy; a ramping SDR needs volume of reps and tight feedback; a will problem needs accountability, not another drill. One template applied to all three wastes the top performer's time and under-serves the new hire.

Rescuing on live calls. Jumping in to save a stuck call on a joint ride-along protects the deal and trains dependence. Unless real revenue is on the line right now, let the rep run it and coach it afterward from the recording.

The misdiagnosis trap. If behavior genuinely changes and conversion does not move within 6–10 weeks, your diagnosis was wrong. Re-examine whether the constraint is skill at all. Common structural culprits: lead quality, ICP drift, a comp plan that pays for the wrong motion, a territory with no addressable accounts, or a product gap the rep cannot talk around. This is where a RevOps partner earns their keep — pulling stage-conversion data by rep, by segment, and by source to separate a coaching problem from a routing or targeting problem.

Recording-consent and trust edge cases. In some regions and industries, call recording carries consent requirements, and in a few enterprise accounts prospects will decline. Build a fallback: shadowed live calls with written notes, or role-play against a scored rubric, so the cadence does not stall when recordings are unavailable. Also watch for the trust failure mode — if reps believe clips are being harvested for performance management rather than development, they will start managing their calls for the recording. Keep coaching clips and performance documentation visibly separate.

Remote and distributed teams. Distributed cadences fail differently: they degrade into camera-off status calls. Countermeasures are shared screen on the clip, cameras on for the diagnosis phase, and a shared doc both people type into live so the commitment is visibly written rather than verbally agreed.

Manager turnover. A cadence tied to one manager's personal habit dies at the handoff. Keeping the commitment history in a shared doc that a new manager inherits — rather than in someone's private notes — is what makes the practice survive a reorg.

A practical rollout plan

Roll it out over four weeks, then run a 90-day checkpoint. Do not launch it as a program with a name and a kickoff deck; launch it as a calendar change and a habit.

Week 0 — set the conditions. Book the recurring blocks for every rep before you announce anything, at times you can genuinely defend. Create one shared coaching doc per rep with two standing sections: "Last week's commitment" and "This week's commitment." Pick and write down the one scorecard you will use for every rep so coaching is comparable across the team. Tell your own manager what you are protecting and why, so the block survives a leadership escalation.

Week 1 — establish the mechanic. Run the first session with a low bar: one clip, one behavior, one written commitment. Do not try to fix everything you have been noticing for six months. Say out loud how the cadence will work, including that you will open every future session with last week's commitment. Expect this session to feel thin. That is correct.

Week 2 — prove inspection is real. Open with last week's commitment before anything else, even if the rep didn't do it. Especially if the rep didn't do it. The team learns more from watching you inspect an unmet commitment calmly than from any content you deliver.

Weeks 3–4 — add practice. Once the loop is habitual, layer in drills: an objection gauntlet where you play the buyer and fire the top three objections for that deal type, three reps each until the language is smooth; a discovery role-play scored against a consistent checklist; a five-minute pre-call rehearsal of the opening 90 seconds before a high-stakes call; and self-scoring, where the rep grades their own call against the scorecard before the session. Self-scoring is the highest-leverage of these because it moves diagnosis onto the rep.

Week 4 onward — hand over ownership. Every fourth week, the rep runs the session: they pick the clip, lead the diagnosis, and propose their own commitment. This is the single strongest protection against cancellation, because the block stops being the manager's meeting and becomes the rep's. It also surfaces what the rep actually thinks their gap is, which is often more accurate than the manager's read.

Day 90 — audit, don't celebrate. Pull three numbers: adherence percentage, inspection percentage, and stage-conversion movement at the coached stage. Then read three commitment docs end to end and ask whether the commitments got more specific over time. Vague commitments trending vaguer means the loop is running but hollow.

One rollout warning: do not start this in the last three weeks of a quarter. Launch in the first month of a quarter so the cadence accumulates four to six clean sessions of habit before the first end-of-quarter push tests it. A cadence that has never survived a crunch is untested; a cadence launched during one is stillborn.

Related questions

How is coaching different from a pipeline review?

Pipeline review is about deals — stage, amount, next step, forecast. Coaching is about the rep's repeatable skill. Blending them means the urgent deal always wins and the skill work never happens. Run them on different days with different prep.

What if the rep resists the weekly cadence?

Diagnose whether it is a trust problem or a will problem. Trust responds to leading with the rep's agenda and asking permission before adding input. A genuine will problem in a fairly resourced rep is a performance conversation, not more coaching.

Can this work without call-recording software?

Yes, with more manager effort. Substitute shadowed live calls with written notes, recorded role-plays, or reviewed email and proposal threads. The requirement is a shared artifact both people can point at — not a specific platform.

How do you scale the cadence past eight reps?

Prioritize weekly 1:1s for the reps where coaching has the most leverage, move the rest to biweekly, and add a weekly group call-review where the whole team diagnoses one recording together. Group review covers breadth; the 1:1 covers depth.

What does RevOps contribute to a coaching cadence?

RevOps supplies the stage-conversion, activity-quality, and ramp-time data that tells you whether coaching is the right lever at all — and flags when the real constraint is territory, comp, lead quality, or ICP fit rather than rep skill.

FAQ

How long should each weekly coaching session be?

Thirty minutes per rep is the practical default — enough for an inspection open, a two-to-three-minute clip, rep-generated options, and a written commitment. Forty-five minutes is better when your span of control allows it. Under twenty minutes the options phase collapses and the manager starts telling instead of asking, which is where skill transfer stops.

What do you do when the quarter implodes and there's no time?

Never cancel, only shorten. Run the identical loop in fifteen minutes: sixty-second clip, one question, one-sentence commitment. Cancelling once during a crunch is the most common single cause of a cadence dying, because it teaches the rep that coaching is what gets dropped when things actually matter.

How many behaviors should you coach at once?

One. Two occasionally, three never. A rep under quota pressure can hold one new deliberate behavior at a time, and a cadence that assigns three ends up inspecting none of them. Narrowing to one behavior also makes the following week's inspection unambiguous — either it showed up in calls or it didn't.

How soon should you expect results?

Behavior change visible in recordings within two to four weeks, movement in the coached stage's conversion rate within six to ten weeks, and quota-attainment impact one to two quarters out. Judging a cadence on quota at week four is the fastest way to kill a program that was working exactly on schedule.

Should the rep ever run the session?

Yes — make every fourth week rep-owned. They pick the clip, lead the diagnosis, and propose the commitment. It converts the block from the manager's meeting into the rep's development time, which sharply reduces cancellation risk, and it reveals what the rep believes their real gap is.

What's the single habit that makes a cadence stick?

Opening every session by inspecting last week's written commitment. Nothing else creates the same compounding effect. Managers who keep only that one habit and drop the drills still get partial results; managers who keep everything else and drop the inspection lose the cadence within a month.

Sources

• HBR — The Right Way to Coach Your Salespeople
• Gong Labs — Sales Coaching Research and Data
• RAIN Group — Sales Coaching Best Practices
• Sandler — A Manager's Guide to Sales Coaching
• MindTools — The GROW Model of Coaching and Mentoring
• Salesforce Blog — Sales Coaching Techniques
• Winning by Design — Resources and Frameworks
• SHRM — Coaching and Performance Management

Related on PULSE

• [How do you build a coaching cadence with a distributed sales team?](/knowledge/cg0183)
• [How do you apply the GROW model in a weekly sales 1-on-1?](/knowledge/cg0929)
• [What specific questions do top sales managers ask during a weekly one-on-one coaching session?](/knowledge/cg0807)
• [How do you structure a weekly sales coaching 1:1?](/knowledge/cg0112)
• [How do you run a weekly 1:1 that actually improves rep performance?](/knowledge/cg0804)
• [How do you run a weekly forecast call without it taking two hours?](/knowledge/cg0108)