A sales QBR is the meeting where a revenue team decides what to change next quarter, using what the last quarter actually did as the evidence. That is the whole job. Most teams run something else: one four-hour all-hands where people take turns presenting numbers, a deck gets filed, and nobody commits to anything. That version is a status report wearing a nice template.
This page is the operating model we run when a Series A or B team asks us to fix theirs. It replaces the single all-hands with four smaller meetings, each with a different audience and a written output. If you only have time to fix one of them, fix the deal review. It is the hour that moves money.
Updated September 2026. This version adds a benchmark section so you can judge your own QBR against published numbers, and removes several statistics the earlier version published without a source.
The four meetings at a glance
| Meeting | Who is in the room | Length | The output |
|---|---|---|---|
| Number review | CRO, finance lead, RevOps | 60 min | One page: what we forecast, what we closed, what drove the variance |
| Deal review | CRO, managers, AEs owning the top 10 deals | 90 min | Two dated next steps per deal, and a corrected forecast |
| Rep 1:1 QBR | Direct manager and one rep | 45 min per rep | One change the rep commits to, one the manager commits to |
| Go-forward plan | CRO, managers, RevOps | 90 min | Three to five operational changes, each with an owner and a date |
Total senior time is about six hours spread over two weeks. The all-hands version most teams run is four hours in one day. Same budget, very different output.
Forecast accuracy at most companies, according to Clari. That is a vendor's figure from a vendor selling forecasting software, so weigh it accordingly, but the direction is corroborated everywhere: the quarterly number is usually wrong and most teams never diagnose why.
Why most sales QBRs change nothing
The dominant pattern looks like this. RevOps spends a week pulling data. The CRO spends two days building slides. Reps spend half a day each on territory updates. Everyone gets on a call for four hours. People present their numbers. The CRO asks soft questions. The meeting ends, a recording goes in a folder nobody opens, and two weeks later the forecast looks identical.
The reason this fails is design, not effort. Most QBRs are built around reporting the past instead of changing the future. They answer "what happened" instead of "what do we do now." Reporting is the input to the meeting, not the meeting itself.
The second failure is the wrong audience. A 12-person all-hands turns into a public performance review. Reps defend their numbers instead of asking for help. An honest conversation about a stuck $400K deal does not happen in front of peers, because the meeting structure makes the truth socially expensive.
A working QBR has three properties. It is small enough that people can be honest. It is structured around decisions, not reports. And it produces a written list of commitments anyone can hold the team to next quarter.
The four QBRs you actually need
There is no single QBR. There are four different conversations that most teams cram into one meeting and do badly. Split them.
If you only run one of these, run the deal review. The theatrical all-hands version can die and nobody will miss it.
Meeting 1: the number review
This is the meeting that should not be a meeting. It is a pre-read. RevOps writes a four-page document a week before. Page one is one chart and three sentences: what we forecast, what we closed, what the variance was driven by. The next three pages are the same breakdown by segment, by source, and by stage. No slides. No presentation. People read it before they walk in.
The meeting itself is 60 minutes of questions with the CRO, the finance lead, and RevOps. Three questions are written at the top of the agenda.
What did we get wrong about last quarter's forecast and why? What is the single biggest assumption baked into next quarter's number that could be wrong? What change to the model do we need to make based on what we just learned?
The output is one page, not a deck. It goes to the board, the leadership team, and the head of every other function. If you cannot write the story of last quarter on one page, you do not understand what happened.
The forcing function is the writing. A slide deck lets you hide behind "macro headwinds." A one-pager makes you say "we missed because three deals slipped in segment X, and we did not see the slippage coming because our stage 3 to stage 4 conversion data was stale by 11 days." If your variance keeps coming from the same place quarter after quarter, the problem is upstream in the model, and the sales forecasting accuracy breakdown covers how to fix the inputs rather than the meeting.
If the QBR cannot be summarized on one page, the meeting was a performance, not a working session.
The one-page output is the forcing function. It is what separates a QBR that changes behavior from a QBR that fills time.
Meeting 2: the deal review
This is the meeting that actually moves money, and the one almost no team runs well.
The format. The CRO, the sales managers, and the AE who owns each of the top 10 open deals for next quarter. Not the top 50. Ten, ranked by weighted pipeline value. Each deal gets seven minutes. That is 70 minutes plus a 20-minute wrap, so it fits in 90.
The seven minutes is structured. Two minutes for the AE to state the deal in MEDDIC or MEDDPICC terms. Three minutes for the room to ask one question each, with rotation enforced. Two minutes for the AE to commit to two specific next steps with dates. If your team does not have a shared qualification language yet, pick one first: the MEDDPICC guide covers how to wire it into the CRM so the deal review reads from fields rather than from memory.
The killer is the questions. They are not "how is it going." They are specific. Who is the economic buyer and have you met them. What is the compelling event and how do you know it is real. Who is championing this internally and what happens to them if it does not close. What is the single most likely reason this deal slips. What is the cost to the buyer of doing nothing.
If the AE cannot answer any of those in a sentence, the deal is not where the forecast says it is. Demote it, adjust the number, move on.
Our own experience of running this format is that the first pass always moves a meaningful share of the top 10 out of the quarter, and that the reps were not lying. They were optimistic. The deal review surfaces the gap between optimism and reality early enough to do something about it. We do not publish a percentage for that effect because we have not measured it in a way anyone else could check, and a made-up number would not help you.
What the public data does support is that timing and buying-group access are the levers. The Ebsta and Pavilion B2B sales benchmarks, built from connected CRM data across hundreds of thousands of opportunities, report that deals which slip lose win rate sharply, and that getting a decision-maker engaged early raises it. Both of those are exactly what a deal review is for: it catches the slip before it happens and it names the missing stakeholder while there is still time to reach them.
For a Series A team running this for the first time, it feels brutal. Reps feel exposed. The fix is to make it clear the goal is to help the deal close, not to grade the rep. The rule we use on the first run: the CRO may only ask questions, not make statements, for the first three deals. It changes the social dynamic immediately.
The data side of this meeting matters as much as the format. If your coverage math is wrong you will review the wrong ten deals, which is the failure the pipeline coverage ratio guide is about, and the CRM work behind it is what we do in CRM and RevOps engagements.
Meeting 3: the rep 1:1 QBR
This is the meeting managers skip first when they are busy and the one they should skip last. The rep 1:1 QBR is where you find out who is going to leave in the next 90 days, who is about to underperform for a reason nobody saw, and who is ready for more responsibility.
The format. The direct manager and the rep. 45 minutes, quarterly. Not the weekly 1:1. A different meeting with a different agenda.
Five sections.
First, the rep's own assessment of the quarter. What worked, what did not, what they would do differently. The manager listens and writes it down.
Second, the manager's assessment. Specific, with examples. Two strengths to keep doing. One gap to close. No more than one, because people can only work on one thing at a time.
Third, the career conversation. Where does the rep want to be in 18 months, what skills get them there, and what is the single thing the company can do to help. This is the question that surfaces flight risk months before it becomes a resignation letter.
Fourth, the commitments. The rep commits to one thing they will change next quarter. The manager commits to one thing they will do to support it. Both in writing. That written commitment is also the raw material for coaching, which is a separate weekly loop covered in the sales coaching guide.
Fifth, the rep asks anything. The company, leadership, the comp plan, the territory. Hard questions. The manager answers honestly or says "I do not know, I will find out by Friday."
On why this is worth 45 minutes per rep per quarter: replacing a departed rep is expensive, and the published estimates put it somewhere between $115,000 and $150,000 per departure once recruiting, onboarding and the ramp productivity gap are counted. Those figures come from vendor blogs such as MarketBetter and SalesHive rather than from an independent study with a disclosed sample, and the ranges quoted across sources vary widely, so treat the number as an order of magnitude rather than a benchmark. The order of magnitude is enough to make the argument: one 45-minute conversation that catches flight risk early is cheap against any figure in that range.
Meeting 4: the go-forward plan
This is the meeting nobody runs, and the reason most QBRs change nothing. After the number review, the deal review and the rep 1:1s, leadership sits down and answers one question. What three to five things will we actually do differently next quarter?
Not "drive more pipeline." Not "improve conversion." Specific operational changes with the shape of these:
Move SDR meeting credit from "meeting booked" to "meeting held with a stage 2 advance," because a large share of booked meetings never advance and the current metric pays for the wrong outcome.
Require every deal above a set value to have a documented champion meeting logged in the CRM before it can move to stage 3.
Retire the two worst-performing outbound sequences and put every SDR on the one with the highest reply rate.
Redraw the territory in the region where one AE holds an unworkable share of the named accounts.
Hire a sales engineer, with the case built from your own demo-to-trial conversion with and without one.
Five changes. Each assigned to one owner with a date. Each with a leading indicator that can be measured weekly. RevOps writes it up, leadership signs off, and it gets reviewed in the weekly forecast meeting for the whole next quarter.
If you do not produce this list, the QBR did not happen. You ran a status meeting and called it a QBR.
What good looks like, in numbers you can check
Most QBR advice, including the version of this page that ran before September 2026, tells you what to do without giving you anything to measure yourself against. Here are the published benchmarks worth holding your own quarter up to, with who published each one and what it is actually worth.
| Benchmark | What it says | Source | How much weight it holds |
|---|---|---|---|
| Forecast accuracy | Most companies forecast less than 75% accurately | Clari | Vendor figure from a forecasting vendor, no disclosed sample |
| Quota attainment | 78% of B2B sellers missed quota in 2025, up from 69% in 2024 | Ebsta and Pavilion GTM benchmarks | Connected CRM data across hundreds of thousands of opportunities, the largest dataset in the category |
| Win rate | Roughly 19% average B2B win rate in the 2025 dataset, sharply lower on larger deals | Ebsta and Pavilion | Same dataset, same caveat: it is their customer base, not a random sample |
| Buying group size | Six to 10 decision-makers shape a complex B2B purchase | Gartner | Independent research firm, the most citable figure here |
| Buyer preference | 67% of B2B buyers say they prefer a rep-free buying experience | Gartner, March 2026 | Survey research, directionally consistent across years |
How to use this in the QBR rather than just reading it. Three of these map directly onto the four meetings.
Forecast accuracy belongs in the number review. Compute yours the boring way: actual closed revenue divided by the forecast you committed at the start of the quarter. Do it for four quarters, not one, because a single quarter tells you nothing about whether your process is biased or just noisy. If the variance runs in the same direction every quarter, the model is wrong, not the quarter.
Win rate by deal size belongs in the deal review. If your win rate collapses above a certain deal size the way the benchmark data does, your top 10 list is the wrong 10 and your qualification bar needs to move, not your activity level.
Buying group size belongs in the deal review too, as a single question. How many of the six to 10 people who will shape this decision have you actually met? On most stuck deals the honest answer is one, and that is the finding, not the forecast.
The buyer preference figure is the one to read carefully rather than act on. It does not mean fewer reps. It means the moments a buyer does want a person have to be worth their time, which is a qualification and preparation problem, and both are decided in the deal review.
The data you actually need for a QBR
Most QBR data prep is wasted. Teams pull 30 reports and use four. Here is the minimum set.
If the data is wrong, the QBR is worse than useless, because you will make decisions on bad signal. The most common reason QBR data is wrong is that reps do not update the CRM. The fix is not a stricter rule. The fix is to make the CRM update the easiest thing the rep does that week, with workflow design and automation that pulls activity from the call recorder, the calendar and email automatically.
We have done this for Series A and B teams on HubSpot, Salesforce and Attio, and the pattern is the same every time. Auto-log activity from connected systems. Use a single deal scorecard field the rep can fill in 90 seconds. Run a Friday hygiene check that flags deals with stale fields. After about a month, the data is good enough to run a real QBR on.
The QBR cadence that works
Once a quarter is wrong. The cadence that works is layered.
Weekly: forecast call. 30 minutes. Top 10 deals only. What moved.
Monthly: pipeline review. 60 minutes. Where is next quarter coming from, what is the gap to plan.
Quarterly: the four meetings above, spread over two weeks. Do not cram them into one day.
Annually: strategy off-site. Two days, different format, different output. Not a QBR.
The common mistake is running the quarterly review only after the quarter closes. By then the data is locked and you cannot change anything. Run the deal review and the rep 1:1s in the last month of the current quarter, not the first month of the next one. You can still influence the close.
What changes when you run this
Four things change, and they change for reasons you can trace rather than because the meeting has better energy.
Forecast accuracy improves, because the deal review surfaces optimistic forecasts before they reach the board rather than after.
Attrition surprises drop, because the rep 1:1 QBR catches flight risk while there is still time to act on it.
The number of operational changes the team actually executes per quarter goes from zero or one to four or five, because the go-forward plan creates a written list that gets reviewed weekly.
Senior time spent on QBR prep falls, because one-page documents replace the 84-slide deck and you stop using a single meeting to do four different jobs.
We are not putting percentages on any of those, and the earlier version of this page did. Those figures came from our own engagements, were never measured in a way an outside reader could check, and several of them were specific enough to imply a rigour that was not there. The honest version is the mechanism, plus the benchmarks above to measure yourself against.
None of this is exotic. It is discipline. Pick a format, use it every quarter, write things down, review the commitments. Most teams do none of this because nobody showed them what good looked like, and the template they copied from a sales blog in 2019 is the one with the 84-slide deck.
Want help redesigning your QBR?
We run a 30-minute QBR teardown for Series A and B revenue teams. You send the last deck, we tell you the three changes that would have the biggest impact on next quarter.
Book a QBR teardown →FAQ
How long should a sales QBR actually be?
The full set of four meetings totals about six hours of senior leadership time, spread over two weeks. The all-hands version most teams run is four hours in one day. Same total time, very different output. The split version produces a written go-forward plan. The single-meeting version produces a deck.
Who should attend a sales QBR?
It depends which of the four meetings. The number review is the CRO, finance and RevOps. The deal review is the CRO, managers, and the AEs who own the top 10 deals only. The rep 1:1 QBR is the direct manager and the rep. The go-forward plan is the CRO, managers and RevOps. No part of this needs the entire revenue org in the room.
What is the difference between a sales QBR and a customer QBR?
They share a name and almost nothing else. A sales QBR is internal: your team reviewing its own quarter and deciding what to change. A customer QBR is external: you and a customer agreeing what value they got and what the next quarter's plan is, usually run by customer success ahead of a renewal. Different audience, different agenda, different output. The customer-facing version is covered separately in the quarterly business review guide.
How do you run a QBR for an early-stage team with no CRO?
Same structure, smaller meetings. The founder or head of sales runs the number review with the finance lead. The deal review covers the top 5 deals instead of 10. The rep 1:1s still happen, because they are the highest-impact hour. The go-forward plan is three changes instead of five. A 10-person team can run the whole thing in about four hours.
What is the difference between a QBR and a sales kickoff?
A QBR looks back at one quarter and decides what to change. It is operational. A sales kickoff launches a new fiscal year or a major plan change. It is motivational and educational, covering product training, comp plan rollout and strategy alignment, typically two days off-site with the whole team. QBRs are working sessions. SKOs are events.
Is the QBR still worth it when most buyers want to self-serve?
Yes, and the self-serve trend is an argument for the deal review specifically. Gartner reports that 67% of B2B buyers prefer a rep-free experience, and separately that six to 10 decision-makers shape a complex purchase. Both point the same way: rep time is scarcer and has to be aimed better. Deciding where to aim it is what the deal review does.
What software do you need to run a good QBR?
Less than you think. Your CRM (HubSpot, Salesforce or Attio) for pipeline and deal data. A conversation intelligence tool for activity and call evidence. A shared doc tool for the one-page outputs and the commitment list. That is it. You do not need a dedicated QBR platform. The discipline is the system, not the software.
How we sourced this update
Two limitations worth stating plainly rather than hiding.
First, none of the domains cited here are reachable from the environment this update was written in. Every figure was established from multiple independent search results that agree with each other, and is attributed to the organisation that originated it. Where the sourcing is weak, the page says so next to the number rather than presenting it as settled. That applies to the Clari forecast accuracy figure and to the rep replacement cost range, both of which are vendor estimates without a disclosed sample.
Second, six statistics that appeared on this page before September 2026 have been removed rather than softened. They were our own engagement observations, published as precise percentages, and none of them had been measured in a way a reader could verify. What we can say about our own practice is now stated as a mechanism, not as a number. The one thing on this page that is genuinely ours is the four-meeting structure and the question lists inside it, which is the part worth copying anyway.
If your QBR has turned into theater and you want to redesign it so it actually moves next quarter's number, book a call and we will walk through what is broken and what to change first.