Sales Pipeline Metrics: The Seven Numbers Worth Reviewing

Walk through a typical revenue dashboard and you will find dozens of tiles, none of which settles the argument in the forecast meeting. Three answers are all a sales leader really needs: whether the volume suffices, whether the mix is right, and whether anything is advancing. Seven numbers cover all three, and your CRM almost certainly holds them already.

Below, each number gets a definition, a calculation and the choice it should force. Leaders and RevOps owners after a brief list, not a data-warehouse build, are the audience.

Outcomes, Not Effort

These measures report outcomes per opportunity: the amount of qualified demand created, its pace and the share that turns into revenue. Calls and emails are effort going in. The numbers below are what that effort should yield, and blending the two kinds is how dashboards bloat.

Pedowitz Group's guide to RevOps reporting puts the governing principle well: every metric needs a governed definition, an owner and a reliable refresh cadence. Keep that test in mind for each number below.

Is There Enough? Two Numbers

1. Opportunities created, split by source

Tally fresh qualified opportunities weekly or monthly, labelling each by where it started, whether inbound, SDR outbound, AE-sourced, partner, referral or expansion. Salesloft's 2024 piece, which draws on a survey by the investor Insight, urges teams to measure contribution by source and go more granular, separating demo requests from events and from paid search. Decision it drives: which route deserves more of the team's time. See sourced versus influenced rules for the crediting.

2. Pipeline coverage

Divide open qualified pipeline by the quota still to be earned, at a fixed date in the period. Decision it drives: build more pipeline immediately, or hold. The multiple that counts depends on your win rate, as the 3x coverage rule explains.

Is the Mix Right? Three Numbers

3. Meeting-to-opportunity rate

Of first meetings held, what share becomes an accepted opportunity? A low rate points at either the quality of booking or the quality of discovery, and the split between them is the decision.

4. Stage-to-stage conversion

For each stage, the share of deals that reach the next. It reveals which stage leaks. The figure is meaningful only when each stage marks something the buyer did, which is the subject of the five-day clean-up.

5. Win rate, cut by segment and source

Wins divided by opportunities that reached a decision. In its velocity guide, HubSpot tells readers to split small, mid-market and enterprise pipelines before doing the sums, since one blended rate masks a strong motion sitting next to a weak one. Salesforce illustrates the basic arithmetic with one deal won out of four, a 25% closing ratio. The decision: how much pipeline each segment needs.

Is It Moving? Two Numbers

6. Stage age

How long a deal has waited where it is now, set against the median wait of deals that went on to win. It tells you which deals to examine or release.

7. Pipeline velocity

Multiply qualified opportunities, average deal value and win rate, then divide by cycle length. HubSpot frames it as the speed at which deals pass through and earn revenue. It tells you which input to work on, a choice explored in the velocity guide.

The Seven on One Page

Metric	Field to pull	Cadence
Opportunities by source	Create date, source	Weekly
Coverage	Open amount, quota remaining	Quarterly plan, monthly check
Meeting-to-opportunity	Meeting held date, opportunity created date	Monthly
Stage conversion	Stage history	Monthly
Win rate	Outcome, segment, source	Monthly
Stage age	Date entered stage	Weekly
Velocity	All four inputs	Monthly

Signs the Numbers Are Not Yet Trustworthy

• Reps carry different definitions of a qualified opportunity.
• The pipeline total climbs while bookings stay flat.
• One blended win rate gets quoted for the whole company.
• Proposals age for quarters without being closed out.
• Many records carry Other or a blank source.
• The pipeline figure changes during the forecast call.

Spot two of these and you are steering by figures you cannot trust yet. Gong's 2022 survey of 928 revenue professionals found that only 24% of sales leaders were confident in their team's forecast, and that 52% said their organisation regularly misses its forecast by more than 10%. It is vendor research and four years old, but the direction matches what most RevOps teams report.

Where the Distortion Comes From

Stages usually describe what the seller did. A demo completed is a seller action; a buyer agreeing to run a security review is a buyer action. Conversion between stages means little unless stages follow buyer commitments. Source is frequently an optional field, and optional fields get skipped, which reduces the first metric to a guess. And averages blend motions, so a healthy small-deal engine can mask a failing enterprise one.

Start With Stage Age, Not Coverage

Most leaders open with coverage, since it feels like the number that predicts the quarter. I would start elsewhere. Coverage is derived: it multiplies your pipeline by an assumption about win rate, so it inherits every flaw in your stage definitions and win-rate data. A coverage figure built on ghost deals looks reassuring right up to quarter end.

Stage age exposes those ghosts directly. It needs only one date per stage, it is hard to argue with, and it flags the deals most likely to be inflating everything else. Review it weekly, and ask for each aged deal what the next buyer action is. Close what has none.

A very small team with a handful of deals can skip this order, since stage age is noise and a simple coverage check is enough. The order also fails when your data has no stage history at all; then the first job is capturing dates, before any metric means anything.

Setting It Up

1. Agree a single meaning for qualified opportunity and enforce it as a checklist when records are created.
2. Require a source from a brief dropdown and lock it once the record exists.
3. Show win rate, cycle and deal size per segment ahead of any blended figure.
4. Fix a cadence in the calendar: creation and stage age every week, conversion and velocity every month, coverage and source mix every quarter.
5. Keep the dashboard to seven. A metric that triggers no decision comes off.

When to Involve RevOps

Call in RevOps or an analyst when stage history is unreportable, when a fifth of deals lack a source, or when two leaders cite different win rates in one meeting. Repair the data before the display; a polished dashboard on bad data merely makes the wrong figure look official. The metrics also sit inside the larger system described in the pipeline generation guide.

FAQs about pipeline metrics

Which pipeline metric matters most?

None wins outright. Begin with opportunities by source, to see where pipeline originates, and stage age, to see which deals deserve doubt.

How many metrics are enough?

Seven or fewer, each tied to a decision someone will make.

How do coverage and velocity differ?

Coverage is a snapshot ratio of open pipeline to the quota left. Velocity is a rate, the revenue the pipeline can generate each day.

How often should the dashboard be reviewed?

Creation and stage age deserve a weekly look. Conversion, win rate and velocity suit a monthly one, and coverage plus source mix a quarterly one.

Should marketing-influenced pipeline be counted?

Yes, but in its own column, clearly labelled and never merged with sourced pipeline. The definitions page has the rules.