RevOps Metrics: Pipeline Coverage, Forecast Accuracy, Win Rate by Source

The dashboard you built last quarter probably has a tell. Either the CRO opens it every Monday and the CFO never references it, or finance pulls numbers off it for the board deck and the CRO quietly disputes the win-rate cut on every Friday call. One of them stopped trusting it. You haven't been told yet.

This is the failure mode of "one dashboard for both." RevOps gets pulled between a CRO who wants pipeline narrative and a CFO who wants unit economics, and the compromise is a 14-tile dashboard where every tile is half-defined and nothing is load-bearing. When the two of them sit in the same room and disagree, the dashboard makes the disagreement worse, not better.

The fix isn't a better tool. It's six metrics that both stakeholders can read the same way, with definitions tight enough to survive a review, healthy ranges grounded in B2B SaaS benchmarks, and a QBR slide pattern that ends the meeting instead of extending it.

Pipeline Coverage

Definition: Open pipeline value with a close date in the current quarter, divided by the remaining quota for that quarter. Recalculated weekly. Rolling, not snapshot.

Formula: (Sum of open opportunity ARR with CloseDate ≤ end of quarter) / (Quota − Closed Won ARR for quarter)

Healthy range: 3-4x at the start of the quarter for B2B SaaS with a 60-90 day average sales cycle. Drift down to ~2.5x by week 10 of a 13-week quarter is normal as deals close. Under 2.5x at the start of the quarter is a miss in the making. Over 5x is not strength. It's stage inflation. Reps are leaving deals open in early stages to pad the number, or marketing is dumping unqualified MQLs into Stage 1 for attribution credit.

Failure mode: Coverage looks great but win rate is dropping. That's the inflation tell. The pipeline grew faster than the conversion rate justifies, which means most of the new entrants aren't real.

CRM source: Opportunity object, filtered to IsClosed = false and CloseDate IN current quarter. Reconcile weekly to a frozen snapshot, a Monday-morning extract that doesn't get rewritten when reps push close dates on Tuesday. Without the snapshot you can't tell whether coverage moved because deals advanced or because dates got slid.

Worked example: Q2 quota is $4M. You've closed $1.2M, remaining quota is $2.8M. Open pipeline with a Q2 close date is $9.5M. Coverage is 3.4x. Healthy. But check the cut by stage. If $6M of that $9.5M is in Stage 1 or Stage 2 with under 30 days to quarter-end, the headline number is lying. Real deployable coverage is closer to 1.6x.

Forecast Accuracy

Definition: Variance between the rep-submitted commit + best case and actual closed-won at quarter-end, measured at fixed checkpoints.

Formula: |Actual Closed Won − Forecasted Number| / Forecasted Number

Healthy range: ±10% by mid-quarter (week 6-7 of a 13-week quarter), ±5% by week 10. Some teams pull tighter (high-velocity SMB SaaS can hit ±3% by week 10), but ±5% is the bar that holds for most B2B SaaS in the $10M-$100M ARR band.

Failure mode: Persistently positive variance (you beat forecast by 15%+ every quarter) is sandbagging. Persistently negative variance (you miss by 15%+) is a process problem, not a rep problem. Either way, the issue isn't that reps are bad at forecasting. It's that nobody is calibrating what "commit" means across the team.

CRM source: Opportunity forecast category (Pipeline / Best Case / Commit / Closed) with a forecast snapshot history. If your CRM doesn't snapshot weekly, build a custom object that does. Without history, you can't tell whether a deal slipped from Commit to Best Case or whether the rep just changed their mind on Friday.

Worked example: Week 7 commit is $2.3M, best case is $3.1M. Actual closed-won at week 13 is $2.7M. Variance from commit is +17%, variance from commit + best case midpoint ($2.7M) is 0%. The team is calling commit conservatively. That's not "good forecasting." It's a coaching problem. Reps are protecting themselves at the cost of giving the CRO a usable number.

Win Rate by Source, Segment, and Rep

Definition: Closed-won opportunities divided by closed (won + lost) opportunities, cut three ways. Always three.

Formula: Won / (Won + Lost), segmented by lead source, customer segment (SMB / Mid-Market / Enterprise), and individual rep.

Healthy range: Highly business-dependent. For mid-market B2B SaaS with inbound + outbound mix, a blended 22-28% win rate is typical. But the blended number is the metric most likely to lie to you. The cuts are where the truth lives.

Failure mode: A single channel collapses inside a stable blended number. Inbound drops from 35% to 22% over two quarters while outbound climbs from 12% to 18%. Blended stays at 26%, the dashboard shows green, and nobody notices that paid acquisition just got 40% less efficient. Or one segment quietly stops converting: enterprise win rate drops from 30% to 18% while SMB wins compensate, and you don't catch it until a six-month enterprise pipeline review.

CRM source: Opportunity object joined to lead source, account segment, and owner. If lead source is multi-touch attribution, pick one model and stay there. Switching from first-touch to last-touch mid-quarter makes the trendline meaningless.

Worked example: Blended win rate is 26%, flat for three quarters. Cut by source: inbound 35% → 31% → 22%. Outbound 14% → 16% → 18%. The flat blend hides a paid-search channel that's degrading fast. Diagnosis: either ICP drift in inbound leads (marketing bought traffic that doesn't fit), or the SDR-to-AE handoff is breaking on inbound specifically. You won't know which until you cut by source AND stage. But you would have missed it entirely if you only looked at the blend.

Sales Cycle by Stage

Definition: Median days a won opportunity spent in each stage, computed from stage-history records (not from CreatedDate to CloseDate).

Formula: For each stage S, median of (DateExitedS − DateEnteredS) across all closed-won deals in the trailing 6 months.

Healthy range: Wildly business-dependent. Enterprise B2B SaaS averages 90-180 days, Mid-Market 45-90, SMB 14-45. The benchmark that matters is your own trailing baseline. If Stage 3 (validation) used to take 12 days and now takes 28, you have a problem regardless of what the industry average says.

Failure mode: Stage 3 → Stage 4 stall. This is the most common silent killer in B2B SaaS. Deals pile up in "Validation" or "Technical Evaluation" because a stakeholder went silent, the rep is afraid to push, and the CRM has no rule that flags it. The deal looks active in pipeline coverage, contributes to forecast, then closes-lost in week 13 with the note "no decision."

CRM source: Opportunity stage history. This is a custom object in most CRMs. If you don't have one, build it before you build the metric. A flat LastStageChange field doesn't give you median time-in-stage across deals.

Worked example: Median time in Stage 3 was 14 days last year, is 22 days this year. Deals exiting Stage 3 to closed-lost grew from 18% to 27%. Diagnosis: Stage 3 entry criteria are too loose. Reps are advancing deals to Stage 3 to claim coverage, then those deals stall because they were never really qualified. Fix is at the entry gate, not the exit. Tighten Stage 2 → Stage 3 criteria (proof of pain documented, economic buyer identified, decision criteria captured), and the cycle compresses.

Comp Attainment Distribution

Definition: The percentile distribution of quota attainment across the rep team. The shape, not the average.

Formula: Sort reps by attainment % at quarter-end. Plot as a histogram or report 10th / 50th / 90th percentile values. Average attainment is on the slide too, but it's the supporting cast, not the lead.

Healthy range: A healthy team has a roughly normal distribution centered around 95-105% attainment, with the 90th percentile at ~140% and the 10th percentile not below ~50%. Roughly 60-70% of reps clearing 80%. That's a team where territories are fair and quota is calibrated.

Failure mode: Bimodal distribution (60% of reps under 80% attainment with two reps over 200%) looks fine on the average (might still average 95%) and is actually a five-alarm fire. It means territories are unbalanced. Two reps got the easy patches and everyone else is fighting for scraps. The CFO sees an average and thinks "fine." You should see the shape and call it.

The other failure mode: tight distribution with everyone clustered at 90-110% attainment. Reads as "well-calibrated" but is often the signal of soft quotas. Quotas that nobody misses badly and nobody beats badly aren't quotas, they're salary justifications. Real sales teams have a long right tail. If yours doesn't, the comp plan is not stretching anyone.

CRM source: Comp/quota object joined to user. If your comp engine is a spreadsheet, the histogram needs to live in BI, not in the comp tool.

Worked example: Average attainment is 96%, looks healthy. Distribution: 4 reps over 150%, 8 reps between 50-75%, 3 reps between 90-110%. That's a broken territory map dressed as a fine quarter. The diagnosis is named: territory rebalancing required, not "rep coaching."

CAC Payback

Definition: Months required to recover the fully-loaded customer acquisition cost from gross margin on new customer revenue.

Formula: (S&M Spend in period) / (New Customer ARR in period × Gross Margin %), expressed in months.

Healthy range: Under 12 months for healthy B2B SaaS. 12-18 months is acceptable for enterprise-heavy motions with longer contract values. 18+ months is a board conversation. Over 24 is a fundraising problem.

Failure mode: CAC payback creeps without anyone flagging it because the metric updates monthly, not weekly, and lives in finance's deck rather than RevOps's dashboard. By the time it's on the agenda, three quarters of unit-economics drift have accumulated.

This is the metric the CFO actually cares about. If you only put one CFO-facing number on your dashboard, this is it. CAC payback links pipeline coverage (the CRO's metric) to capital efficiency (the CFO's metric) through win rate, ACV, and gross margin, which is why all six of these metrics need to be on the same page. Move one and you'll see it ripple to CAC payback within a quarter.

CRM source: RevOps doesn't own the inputs alone. S&M spend comes from finance, gross margin from finance + product, new ARR from CRM. Build the metric collaboratively or it'll get disputed at the QBR.

Worked example: Q1 S&M spend $3.2M. New customer ARR booked Q1 $2.4M, gross margin 76%. CAC payback = 3,200 / (2,400 × 0.76) = 17.5 months. Last year same quarter: 13.2 months. Diagnosis: either ACV dropped, win rate dropped, or marketing spend grew faster than pipeline produced. Cross-check against the win-rate-by-source cut and the segment mix. If enterprise win rate fell, that's your CAC payback story.

The "Low Coverage But Everyone Hits Quota" Diagnostic

This pattern shows up about twice a year on a healthy team and it always means the same thing: reps are sandbagging pipeline to protect commit.

The shape:

Pipeline coverage trending down quarter-over-quarter (4.1x → 3.6x → 3.0x → 2.7x).
Attainment trending up or holding steady (98% → 102% → 105%).
Forecast variance staying tight (±5%).
Time-in-late-stage shrinking (deals appear and close inside 30 days).

That's not magic. Reps are working deals offline, in their notebooks, and only entering them in the CRM when they're 80% sure they'll close. They've learned that pipeline you create early gets micromanaged, and pipeline you create late gets celebrated. The system rewards the wrong behavior.

How to confront it without nuking trust:

Don't accuse anyone. The behavior is rational given the incentives.
Run a one-month "pipeline amnesty": every deal entered before Stage 3 doesn't count against the rep in 1:1s. Just track that it exists.
After the amnesty, expect coverage to spike from 2.7x to maybe 3.8x. That's the real number.
Then reset the comp + leadership conversation around what the early stages are for. They're for forecasting, not for performance management.

If your team has this pattern and you fix it, your forecast accuracy will get worse for one quarter (because there's now real pipeline to predict against, with realistic conversion rates) and then get materially better. The CRO will hate quarter one of the fix. Stay the course.

Vanity-Metric Traps to Cut From the Dashboard

MQL → SQL conversion in isolation. Useful as a marketing operations metric. Useless as a revenue metric without the downstream cut to closed-won. A team can hit MQL targets, hit SQL targets, and miss revenue by 20%, and the dashboard celebrates marketing. Always tie MQL → SQL to SQL → Closed Won by source.

"Pipeline Created" without source attribution. A pipeline-created number on its own tells you reps were busy. It doesn't tell you whether they were busy on the right things. Always cut by source and segment.

Activity metrics dressed as outcomes. Calls made, emails sent, demos booked. These belong on a sales-ops operational dashboard for the manager, not on the RevOps QBR slide for the CRO and CFO. The board doesn't care how many calls got made; they care whether those calls converted.

Logo count without ARR cut. "We added 47 logos this quarter" can hide a 30% drop in average deal size. Always show new logos AND new ARR AND average ACV together, or don't show logos at all.

The QBR Slide Pattern

One slide per metric. Same layout every time. Six slides for six metrics, plus a one-page summary. No chart soup, no 14-tile dashboards, no "let me walk you through this real quick." Every slide should be readable in 30 seconds.

The layout, repeated for all six:

─────────────────────────────────────────────
  METRIC NAME                        Q2 2026
─────────────────────────────────────────────
  CURRENT VALUE                  TREND (4Q)
  3.2x                          4.1 → 3.6 →
                                 3.0 → 3.2

  HEALTHY RANGE                  STATUS
  3.0 - 4.0x                    GREEN

  DIAGNOSIS
  Coverage recovered after Q1 territory
  rebalancing. Stage-2-to-3 advancement
  rate normalized.

  ACTION (ONE ITEM)
  Hold current stage criteria. Re-review
  in Q3 if coverage drifts under 2.8x.
─────────────────────────────────────────────

Five elements per slide, no more:

Current value: one number, big.
Trend: last 4 quarters, sparkline or arrow.
Healthy range: the band, not a single target.
Diagnosis: one sentence naming what's happening, not "monitor closely."
Action: one thing being done. Not three. One.

If you can't fit a metric into this layout, the metric isn't ready for the QBR. Either the data isn't trustworthy yet or the diagnosis isn't sharp enough. Cut it from the slide and put it back when it's ready.

Closing

The dashboard's job is to end an argument, not start one. If the CRO and CFO read the same six numbers and reach the same conclusion about what to do next, the dashboard works. If they read it and disagree, either a metric is wrong, a diagnosis is missing, or one of them is reading a different number than you think.

Most RevOps dashboards fail because they try to be exhaustive. The good ones are short, opinionated, and named. Six metrics. Healthy ranges grounded in B2B SaaS reality. A diagnosis on every slide. One action per metric. That's the whole job.

Build that, and you stop being the person who sends out the dashboard. You become the person who reads it back to the room and tells everyone what it means. That's the difference between a RevOps Manager and a Director of RevOps, and it's mostly a question of how confidently you can name what's happening on each line.

RevOps Playbooks