# Managing sales managers: what a sales director should check in the CRM each week https://salesbond.app/blog/managing-sales-managers-crm-pipeline-review/ Published: 2026-10-20 · Updated: 2026-10-20 Managing sales managers comes down to one weekly habit: look at the same five CRM slices for every team (stale deals, deals without a next step, stage spread, forecast accuracy and lost reasons) and coach each manager on how they review their reps, instead of reviewing the deals yourself. Most guides on managing sales managers are really about managing reps; this one is about the layer above. You have three or four managers, each with their own pipeline, and you need to know which one needs you this week without opening every deal. ## Managing sales managers: the job changes more than the title The first time you manage managers, the instinct is to keep doing what got you promoted. You were good at reading a pipeline, so you read all of them. You were good at unsticking deals, so you unstick deals in every team. For a quarter it even works, because you are still the best deal reviewer in the building. Then it stops working. Quietly. The managers learn that you will look at everything, so they look at less. The reps learn that the director will step in on big deals, so they escalate instead of working them. And you have no time left for the thing only you can do, which is to notice that Team B’s forecast misses quarter after quarter while Team A’s lands, and ask why. We think the honest description of the job is this: a manager runs deals through reps, and a director runs pipeline reviews through managers. Your unit of work is not a deal. It’s a manager’s weekly review, and whether it produces a clean pipeline and a believable forecast. That is also what separates a strong director from a busy one: they judge a manager by the quality of the review, not by the deals they happened to hear about. Everything below follows from that. ## Why your view cannot be a deal-by-deal view A director who reviews deals becomes a second manager, and a worse one. You know less about each deal than the person who owns it. You see the CRM record, not the last call. So when you open a deal and ask “what is happening here”, one of two things follows. Either the manager answers, and you have spent ten minutes learning what they already knew. Or the rep answers, and the manager has just been bypassed in front of their team. There is a second cost and it shows up in the data. If the director is known to read deals, the deals get written for the director. Notes get longer and vaguer. Close dates get pushed to the month that keeps the deal out of this week’s conversation. We have [written about the CRM data quality metrics](https://salesbond.app/blog/crm-data-quality-metrics/) that catch this from the manager’s side; from the director’s side the fix is the same. Stop reading the deal. Read the shape of the pipeline the manager is producing, and talk to the manager about the shape. The shape is five numbers. ## Five slices, the same five for every team ![Table of five CRM slices with what each shows, and empty rows for Team A, B and C.](https://salesbond.app/wp-content/uploads/2026/10/managing-sales-managers-crm-pipeline-review-2.png) Pick five filters, build them once, and run them on every manager’s pipeline with identical settings. The settings matter more than the choice of five. Say one team’s stale rule is set to 14 days and another’s to 30: the comparison between them is noise, and the manager with the softer rule will always look better. 1. Stale deals: filter open deals with no activity logged for longer than your agreed limit, as a share of the team’s open deals; it shows whether the manager’s review actually touches old deals or only the ones reps bring up. 2. No next step: filter open deals with no scheduled next activity and no owner-set next action; it shows whether a review ends with every deal having a next move, or just with a feeling. 3. Stage spread: count the number and value of open deals per stage and put it next to the same picture four weeks ago; it shows whether deals move or pile up, and at which stage the team loses momentum. 4. Forecast accuracy: compare the manager’s committed number at the start of each period with what actually closed, over the last three periods; it shows whether the manager reads the pipeline or repeats what reps tell them. 5. Lost without a reason: filter closed-lost deals in the period with an empty or “other” loss reason; it shows whether the manager cares about why, or only about the number. How do you read them? As a share, not a count. A big team will always have more stale deals than a small one, so share of open deals and share of closed-lost is what you compare across managers. Trend before level. A manager whose stale share has been falling for several weeks is doing the job. A manager whose share is low but creeping up is the one to talk to, even if the number still looks fine this week. And all five together. One high slice is a habit to fix. Three high slices on the same team is a manager who is not running reviews at all, whatever the calendar says. That is a different conversation, and a more urgent one. If you want the fuller picture of what a first-line manager’s own weekly checks should cover, the [sales manager responsibilities article](https://salesbond.app/blog/sales-manager-responsibilities-crm-weekly-checks/) has them; the point of this page is that you check the manager, not the checks. ## Thirty minutes a week, per manager, with the numbers open The meeting is short on purpose. If it needs an hour, you have slipped into reviewing deals. Open with the five slices for that team, this week against last week. Don’t ask “how is the pipeline”. Ask “stale share went from here to here, what did you do in your review this week”. The second question has an answer about the manager’s work. The first has an answer about the reps, and the meeting drifts. Then one deep-dive, and only one. Pick the slice that moved the most or sits furthest from the other teams. Say Team C has had the highest no-next-step share for a few weeks running. You do not open those deals. You ask the manager how thier review handles a deal with no next step. Do they ask the rep to set one in the meeting? Do they let “I’ll follow up” count? Do they check the next day that it was set? Close with one commitment from the manager that you can see in the CRM next week. “I’ll push harder on next steps” is not one. “Every deal in the proposal stage will have a dated next activity by Thursday” is, because the no-next-step filter will show it or not. That is the whole meeting: one slice, one question about the review, one commitment. Why weekly and not monthly? Partly because MySalesCoach’s [coaching research](https://www.mysalescoach.com/sales-coaching-statistics) finds that reps coached weekly hit quota far more often than reps coached quarterly or less. That finding is about reps and their managers. We think it holds one level up for a simpler reason: a review that happens every week catches drift when it is a week old, and a monthly one catches it a month late. A disclosure, since this is where tooling comes up: we build SalesBond, a sales management layer on top of the CRM a team already uses; we started with Pipedrive. Its Sales Dashboard has a Team Performance tab with per-rep execution, and Deal Health checks the next step, close date and stage on every open deal against more than 20 rules. [Check your CRM with SalesBond](https://salesbond.app/) if you’d rather start the 1:1 from that view than rebuild filters every Monday. Filters aren’t the hard part, though. Coaching the manager’s review is. ## Coach the review, not the deal Here is the test. After a 1:1 with a manager, ask yourself what changed: a deal, or the way that manager runs their Tuesday pipeline meeting. If the answer is a deal, you did the manager’s job for thirty minutes, and the review they run next week will be exactly as good as the last one. What a manager’s review looks like is visible in the CRM a day later, which is why you can coach it without sitting in the room. A good review leaves every discussed deal with a dated next activity, a stage that matches the evidence and a close date someone would bet on. A weak review leaves the deals as they were, plus a note. You can tell the two apart from the no-next-step slice and from how many deals changed stage the day after the meeting. We cover the reps’ side of this in [sales coaching from CRM data](https://salesbond.app/blog/sales-coaching/); the director’s version is the same data read one level up. The second thing to coach is expectations. MySalesCoach’s [guide to managing a sales team](https://www.mysalescoach.com/blog/sales-team-management) makes the point that one of the most common reasons reps underperform is ambiguity, and that a manager should spell out early how the CRM is used and what a good week looks like. Most managers who run poor reviews have not done this. They inherited a team, assumed everyone knew the rules, and now run a review where half the reps think a next step is optional. Ask the manager plainly: have you told your team what a complete deal record looks like? If they have to think about it, that is the week’s commitment. The third is the forecast. Ask the manager to walk you through how they got to their committed number. If the method is “I asked each rep and added it up”, the forecast accuracy slice will keep telling you so. A manager who reads the pipeline will have left out deals with no next step and deals that have slipped twice, and will be able to say which ones. The [six-number dashboard](https://salesbond.app/blog/sales-manager-dashboard-six-numbers/) we described for managers is a reasonable starting point for what they should look at before they commit a number to you. ## Two mistakes we see most The first is rescuing. A big deal is wobbling in Team A, the manager is unsure, and you take it. You know the buyer’s CFO, or you have closed this kind of deal before. Sometimes you even win it. What you have also done is tell the manager that the hard ones are yours, and tell the the rep that the manager is not the person to go to. The next wobbling deal arrives on your desk faster. If you must be in a deal, go in as the manager’s resource: the manager stays the owner in the CRM, runs the plan, and brings you in for one call. The record should show that, and so should the room. The second is letting each team keep its own rules. Team A logs activities, Team B uses notes. Team A marks a deal lost with a reason, Team C uses “other” for everything. Each manager will defend their version, and each version is probably fine on its own. Together they make the five slices unreadable and your weekly report to the owner a guess. One rule set across teams is a director’s decision, and one of the few that only a director can make. The [player-coach article](https://salesbond.app/blog/player-coach-sales-manager-pipedrive/) describes how a single manager drifts into working deals instead of running the team; with several managers the drift multiplies unless the rules are shared. There is a smaller third one: reporting upward what the managers told you instead of what the CRM shows. If you write a weekly note to the owner, [what to include and what to cut](https://salesbond.app/blog/weekly-sales-report-what-to-include-and-what-to-cut/) applies, and the five slices per team make a better spine for it than a paragraph per manager. ## Monday check Before the first 1:1 of the week, make sure the five filters run with the same settings on every team, because a different stale limit on one team makes the whole comparison worthless. Look at last week’s commitments in the CRM rather than asking about them. And if a deal is open on your screen, close it and ask which manager’s review should have caught it. If every manager leaves their 1:1 with one commitment you can see in the CRM by Friday, you are managing managers. If they leave with a list of deals you want updates on, you are still managing reps, with extra steps.