How do you coach a rep who won't keep the CRM updated?

The two paths: fix the human or fix the friction

Every manager facing a rep who won't keep the CRM updated is really choosing between two interventions, and picking wrong burns four to six weeks.

Path A — coach the human. You treat the gap as a behavior problem. You run structured 1:1s, you connect CRM hygiene to the rep's own win rate, you set an agreement with a review cadence, and you escalate to a documented consequence if the agreement breaks. This is the right path when the rep *can* log a call in under 45 seconds but doesn't, when their peers on the same instance with the same quota are at 90% field completeness, and when the gap is selective — they log the deals they're proud of and skip the ones that are slipping.

Path B — fix the friction. You treat the gap as a system problem. You audit required fields, kill the ones nobody reports on, wire auto-logging from the dialer and calendar, and reduce the number of places a rep has to type the same fact. This is the right path when *everyone* on the team is stale, when logging one call takes more than a minute, when the rep is typing the same customer name into three tools, or when field reps have no usable mobile path.

The trap is that Path A feels like management and Path B feels like someone else's job, so managers default to A. That's backwards more often than not. If three of your seven reps are below the line, you do not have three willpower problems — you have one workflow problem wearing three hats. The tell is *variance*: low variance across the team points at the system; high variance with one or two outliers points at the individual.

There's a third path worth naming because it masquerades as the other two: fix the ask. Sometimes the data you're demanding genuinely isn't worth the minutes. If you require a "competitive intel" free-text field that nobody has read in a year, the rep skipping it is exercising good judgment, not bad discipline. Before you coach anyone, open your own reporting and ask which fields actually drive a decision — a forecast call, a resourcing choice, a renewal play. Fields that drive nothing get deleted, not enforced.

These paths are not mutually exclusive, but the *sequence* matters enormously. Fix friction first, then coach — because coaching against a broken workflow destroys your credibility. The rep does the thing you asked, discovers it still takes eight minutes, and now they've learned that you don't understand their day. You've spent trust to buy a week of compliance.

How to decide between them

Run the diagnosis in a single 1:1, not over a month of hallway comments. Pull three artifacts before the meeting: the rep's last 30 days of logged activities, their open opportunity list with field completeness, and — critically — the same two things for your top performer as a control. Then ask, in this order.

First, the control question. Is the top performer also stale? If yes, stop. You have a system problem and coaching the weak rep is scapegoating.

Second, the speed question. "Walk me through logging yesterday's discovery call. Share your screen." Time it. Under 45 seconds and clean means skill isn't the issue. Over 90 seconds, or if they open three tabs, you've found friction and you have the receipt to escalate to RevOps.

Third, the selectivity question. Look at *which* deals are stale. If the stale ones cluster on late-stage deals that are slipping, that's avoidance — the rep doesn't want the pipeline review conversation. That's a will issue with a fear component, and the fix is psychological safety before it's accountability.

Fourth, the value question. "What do you think I do with this data?" If the answer is "you check up on me," you've got a knowledge gap about purpose, and the fix is showing them a decision you made with their data that helped them.

The diagnosis takes twenty minutes and saves you a month. Write the conclusion down — literally one line in your 1:1 notes: "Marcus: friction, 2m14s per call log, no dialer sync." That line is what you hand RevOps, and it's what stops you from re-diagnosing the same rep in six weeks.

The numbers that make the case

Vague appeals to "hygiene" lose to concrete math, and reps are good at math about their own time.

The time budget. Ask the rep to estimate minutes per day on CRM entry, then measure it for one week. Most reps who claim "an hour a day" are actually spending fifteen to twenty minutes — the perception gap is the real problem, and showing them the measurement often resolves more resistance than any speech. But when the measurement confirms them, you now have an escalation with a number attached: "This costs Marcus forty minutes a day. Across seven reps that's a workweek every week."

The completeness threshold. Pick a small number of fields and enforce those absolutely rather than twenty fields loosely. Stage, close date, amount, next step, and last meaningful contact is a defensible core. Set the bar at near-total completeness on the top handful of open deals rather than a percentage across everything — a rep can hit 80% overall while every deal that matters is blank.

The staleness clock. Time since last update is a better signal than field completeness. A deal with every field filled and no touch in three weeks is more dangerous than a sparse deal updated yesterday, because the full one looks healthy in the forecast roll-up. Set a threshold — anything past two weeks with no activity gets flagged — and make the flag automatic so it isn't you personally nagging.

The lag target. Time between the call ending and the note existing. Same-day is the practical bar; anything logged more than 24 hours later is reconstructed from memory and quietly worthless. Details decay fast — the specific objection, the name of the person who went quiet, the number they pushed back on.

The cost of being wrong. Frame it in deals, not dashboards. Walk through one deal that died where the missing note was material: no logged champion change, so nobody caught that the buyer left; no logged competitor mention, so no differentiation play. One real example from the rep's own history beats every statistic you could quote, and it's the moment the conversation stops being about compliance.

The comp question. Tying CRM completeness to variable comp works, but it's a blunt instrument and it's the last move, not the first. If you do it, weight it small, define it with a query the rep can run themselves, and give a full quarter of warning. A hygiene gate the rep can't self-audit is a trap, and reps treat traps accordingly — they'll optimize for the metric and stop giving you the qualitative detail that made the notes useful.

What the fix looks like week by week

Sequencing is where most of these efforts fall apart. Managers try to do everything in week one, the rep experiences it as an ambush, and by week three both parties are avoiding the topic.

Week zero — you fix your own house. Before any coaching conversation, audit the fields. Open the reports you actually run and list the fields they read. Every required field not on that list gets proposed for deletion. Check whether the dialer, calendar, and email are syncing; a broken sync is invisible to you and infuriating to the rep. Verify mobile works if anyone sells in the field. This week has nothing to do with the rep, and skipping it is why so many of these efforts read as hypocritical.

Week one — diagnose and agree. One 1:1, thirty minutes. Run the four questions. Land on a written agreement with a specific behavior, not a vague commitment: "Every call gets a note and a next step before I leave the parking lot" beats "I'll do better." Ask them to name the trigger — end of call, end of block, end of day — because behavior attached to an existing trigger sticks and behavior attached to willpower doesn't. Then ask what would get in the way, and remove that thing yourself if you can.

Week two — practice, don't lecture. Screen-share and have them log a complex update live while you watch. You are looking for fumbles, not compliance. If they hunt for a field, that's a skill gap you can close in ninety seconds with a shortcut. If they type the account name manually because search is slow, that's a system finding. Do a deal autopsy on one lost opportunity: open every logged activity, ask what's missing, and connect each gap to a coaching moment you couldn't have.

Weeks three and four — shift the frame. Stop asking about the CRM. Instead, run pipeline reviews strictly off what's in the system and refuse to accept verbal updates. "I only see what's logged" is a policy, not a punishment, and it converts CRM hygiene from an admin chore into the rep's own interest — their deal doesn't get help if you can't see it. This single move does more than any number of coaching conversations, because it changes the payoff rather than the pressure.

Weeks five and six — measure and decide. Pull the same three artifacts from week one. Real improvement means the behavior held for two consecutive weeks without a reminder from you. If it improved only in the days after each 1:1, the behavior is attached to your attention rather than to a trigger, and you need to rebuild the habit anchor. If nothing moved and the friction is genuinely gone, you're now in performance-management territory, and the paper trail you built in weeks one through four is exactly what makes that conversation fair.

The adjacent lesson: this same sequence works for any adoption problem, not just CRM. Sequence tools, forecast-call prep, enablement content, a new qualification framework — the pattern is identical. Fix friction, make the desired behavior the path of least resistance, then attach it to an existing trigger, then hold the review honest. Managers who internalize this stop running one campaign per tool and start running one operating rhythm.

Where RevOps owns the problem, not the manager

A meaningful share of what looks like a coaching problem is an architecture problem, and the manager can't fix it alone. Knowing the boundary keeps you from coaching someone into a wall.

Field sprawl. Required fields accumulate because every request to add one is small and no one owns removal. The discipline is a standing review — quarterly is enough — where each required field must be defended by naming the report or decision it feeds. Undefended fields get demoted to optional or deleted. Expect to cut a third the first time.

Double entry. If a rep logs a call in the dialer and again in the CRM, or updates a stage in the forecasting tool and again in the CRM, you have created a compliance problem out of thin air. Every duplicate entry point is a place a rational person stops. RevOps owns the integration; the manager owns escalating it with evidence — which is exactly what the timing exercise produces.

Validation rules that punish honesty. A rule requiring a close date within the quarter teaches reps to enter a fake close date. A rule blocking stage advance without a filled-in champion field teaches reps to type "TBD." Any rule that makes the honest answer harder than the dishonest one will produce dishonest data, reliably, and no amount of coaching overcomes an incentive that clean.

Automated capture. Conversation intelligence and email/calendar sync can carry most of the mechanical load — attendees, timestamps, follow-ups. What automation can't produce is judgment: why the deal moved, what changed in the buyer's world, what you'd do differently. So the coaching ask shifts from "log everything" to "log what only you know." That's a smaller ask and a more honest one, and reps accept it far more readily than a blanket mandate.

Read-back value. The most underrated fix is making the CRM give something back. A rep who opens their pipeline and sees a genuinely useful prioritization — deals at risk, deals gone quiet, next steps due today — has a reason to keep it updated that has nothing to do with you. Systems people maintain are systems that serve them. If your CRM is purely an input device for management reporting, you will fight this battle forever, and you will keep losing it one rep at a time.

Related questions

What if the rep is your top performer?

Same diagnosis, different leverage. Don't lead with compliance — lead with risk: their deals are invisible in the forecast, and if they're out for a week nobody can cover. Ask them to define the minimum viable record they'll commit to, and hold that.

How is this different for a rep with under six months tenure?

New reps almost never have a will problem — they have a knowledge problem about what "good" looks like. Show them a strong rep's opportunity record as a worked example. Modeling beats instruction here, and the habit sets in the first ninety days or not at all.

Does this change for a remote or hybrid rep?

Only the observation method. You lose hallway awareness, so screen-share becomes essential and a short recorded walkthrough of the exact workflow pays for itself. Set one shared block per week for pipeline cleanup so it's a team ritual, not surveillance.

What if the CRM data is bad from before they got there?

Declare amnesty on the backlog. Nobody cleans a hundred stale records, and asking them to guarantees failure. Clean forward only — new activity, current open deals — and archive or close-lost anything untouched past a threshold.

How do I coach this without damaging trust?

Never raise it in a team meeting, and never raise it only when a deal dies. Praise clean pipelines out loud, and make your first CRM conversation of any month a helpful one — a deal you spotted and can help with — not a corrective one.

FAQ

How long before I should expect to see change?

If the root cause is skill or knowledge, improvement usually shows within a week or two, because the barrier was just not knowing. Will issues take longer — plan on four to six weeks of consistent cadence before you judge, and require two consecutive unprompted weeks before calling it fixed. System issues don't improve at all until the system changes, no matter how long you coach.

The rep says "I don't have time." How do I respond?

Take it literally and measure it. Ask them to time themselves for one week — every minute spent on CRM entry. Either the number is small, and you've resolved a perception problem with data, or the number is large, and you now have an escalation to RevOps with evidence instead of a hunch. Never argue about the estimate; measure it.

Should CRM completeness affect compensation?

Only after coaching and system fixes have both been genuinely tried, and only with warning, a small weight, and a definition the rep can self-audit. The risk is real: gate comp on a metric and you'll get the metric, often at the expense of the qualitative notes that made the record useful in the first place. It's a last resort, not a policy default.

What if the whole team is bad at this, not one rep?

Then it isn't a coaching problem and treating it as one will cost you credibility with everyone at once. Team-wide staleness is an architecture signal: too many required fields, missing integrations, duplicate entry, or a rule set that punishes honest answers. Fix that first, publicly, so the team sees you removed friction before you asked for behavior.

Can automated call logging solve this entirely?

It solves the mechanical half — who attended, when, what was said, what follow-up was promised. It cannot produce the judgment layer: why the deal shifted, what the buyer's real constraint is, what you'd change. So automation shrinks the ask rather than eliminating it, and the coaching conversation becomes about the small set of things only the rep knows.

How do I keep it from sliding back after six weeks?

Anchor the behavior to a recurring event rather than to your attention. Pipeline reviews run strictly off logged data, a standing weekly cleanup block, and an automatic staleness flag do more than any reminder you personally send. If the behavior decays the moment you stop watching, it was never a habit — it was compliance, and compliance always has a half-life.

Sources

• Harvard Business Review — What Good Coaching Looks Like
• Gartner — Sales Insights and Research
• Salesforce — CRM Adoption Resources
• HubSpot Sales Blog
• Gong Labs Research
• MIT Sloan Management Review
• SHRM — Performance Management
• McKinsey — Growth, Marketing and Sales Insights

Related on PULSE

• [How do you coach a sales rep who won't prospect?](/knowledge/cg0021)
• [How do you know when coaching won't fix a sales rep?](/knowledge/cg0138)
• [How do you coach reps to keep their CRM clean and current?](/knowledge/cg0102)
• [How do you coach reps to keep their pipeline 3x covered?](/knowledge/cg0023)
• [How do you keep remote reps accountable without micromanaging?](/knowledge/cg0187)
• [How do you keep coaching consistent when you're swamped?](/knowledge/cg0116)