Pipeline scoring is the practice of scoring every open deal on evidence that it will close, then adding those scores up to show how much of the pipeline is real. The result is a pipeline score, the share of open value that passes, and scored coverage: passing value divided by the target. Updated November 2026: added the pipeline score formula, an illustrative worked example and a comparison with raw coverage.
This page covers scoring the pipeline as a whole. For how a single deal score is built and tested, see predictive deal scoring, what is deal scoring and the deal scorecard.
What Pipeline Scoring Is
Pipeline scoring works at two levels. Each open opportunity gets a deal score from objective signals and buyer behavior. Those scores then roll up into a pipeline score, which says what share of the pipeline is live, and scored coverage, which says whether the live part covers the number.Stage probabilities give a baseline conversion estimate. Deal scores separate a strong deal from a weak one at the same stage. The roll-up answers the question a raw pipeline coverage ratio cannot: of everything in the CRM, how much is likely to turn into revenue? Scoring improves forecast accuracy because it finds deal risk before it becomes a miss.
Pete Furseth, ORM's COO, puts the coverage problem plainly. Treating pipeline coverage as an answer is the "best practice" he ignores: "It is a useful input, but it should never be the conclusion." A team can show 4x coverage and still miss when that pipeline sits in the wrong segment, is too old, or carries amounts deals rarely close at.
How do you calculate a pipeline score?
Score each deal, flag stale deals, then divide the value of deals that pass your threshold by total open value. Scored coverage is that passing value divided by the remaining target.`Pipeline score = value of open deals at or above the pass line / total open pipeline value x 100`
`Scored coverage = value of open deals at or above the pass line / remaining target`
Worked example (illustrative numbers). A team has a $1,000,000 target for the quarter and $3,500,000 of open pipeline with close dates in the quarter, so raw coverage is 3.5x. The pass line is a deal score of 40.
| Band | What puts a deal here (illustrative rule) | Open value | Share of pipeline | Counts toward scored coverage |
|---|---|---|---|---|
| Strong (70 to 100) | Evidence on most signals, a stage or close-date change in the last few weeks | $900,000 | 26% | Yes |
| Live with gaps (40 to 69) | Moving, but missing an executive sponsor, budget confirmation or a dated next step | $1,100,000 | 31% | Yes |
| Weak (under 40) | Single-threaded, no next step, close date already pushed | $800,000 | 23% | No |
| Stale | No change in stage, close date or amount in 12 months | $700,000 | 20% | No, whatever its last score |
| Total | $3,500,000 | 100% | $2,000,000 passes |
The stale row follows ORM's rule for most customers: a deal with no meaningful activity in 12 months stops being real pipeline, and meaningful activity means a change in stage, close date or amount. Pete puts stale deals at 10% or more of the pipeline, varying by customer. More on measuring it in stale pipeline.
Pipeline score vs coverage vs weighted pipeline
Each measure answers a different question, and only scored coverage asks whether the real part of the pipeline is enough.| Measure | Formula | Question it answers | What it misses |
|---|---|---|---|
| Raw coverage | Open pipeline / target | Is there enough pipeline on paper? | Age, quality and amount realism |
| Weighted pipeline | Sum of deal value x stage probability | What is the pipeline worth if every deal at a stage behaves alike? | Differences between deals at the same stage |
| Deal score | Weighted signals per deal, 0 to 100, or a model probability | Which deals need attention this week? | The total picture |
| Pipeline score | Passing value / open value | How much of the pipeline is real? | Whether the real part is enough |
| Scored coverage | Passing value / remaining target | Does the real part cover the number? | Deals that will be created and closed inside the quarter |
How is pipeline scoring implemented?
Each deal is scored on weighted engagement, process and qualification signals, and the strongest signals describe change. As Pete puts it: "For forecasting, the state of the pipeline matters. The movement in the pipeline often matters more." A deal in Proposal is useful information. A deal that moved into Proposal yesterday is more useful, and so is an amount that just went up or down late in the cycle.A simple rule-based model uses three groups. The 40/30/30 split below is an illustrative starting point; your own win and loss history should set the weights.
Engagement signals (40% of score) - Number of buyer stakeholders engaged (multi-threaded vs. single-threaded) - Executive engagement (VP+ level active in conversations) - Meeting frequency and recency - Champion activity (internal advocacy observed) Process signals (30% of score) - Time-in-stage vs. median (deals progressing at normal pace score higher) - Recent change in stage, close date or amount, and its direction - Next steps defined with dates - Mutual action plan in place Qualification signals (30% of score) - ICP fit score - Budget confirmed and allocated - Competitive position (sole vendor vs. competitive evaluation) - Compelling event identified (deadline that creates urgency)A deal scoring 38/40, 28/30 and 25/30 totals 91 and lands in the strong band. One scoring 10/40, 12/30 and 10/30 totals 32 and stays out of scored coverage until something changes. Teams with enough history can replace the hand-set weights with a model; predictive deal scoring covers how that model is built and checked.
Why pipeline scoring matters for revenue teams
Coverage can look fine while deals stop moving. Pete saw this across ORM customers in Q1 and Q2 of 2026: on paper they had enough pipeline, but fewer deals were being decided. "They were not being won. They were not being lost. They were just sitting there." Days in stage crept up and sales velocity slowed while the coverage ratio held. A pipeline score built on movement signals falls in that situation, and raw coverage stays where it was. Reps lose selling time to deals that will never close. When every deal has a visible score, reps can work the deals likely to close and disqualify the ones consuming time without progressing. Pete's advice when decisions slow down is to force them: "A no today is better than a no three months from now."Scoring also changes pipeline reviews. Instead of reviewing deals alphabetically or by size, managers can start with high-value deals in the middle band, where coaching moves the most. High-score deals need less attention. Low-score deals may need to be disqualified.
How to build a pipeline scoring model
- Start with historical data. Pull your closed-won and closed-lost deals and find which signals separated wins from losses. Give the strongest signals the most weight. For a machine learning model, Pete wants a couple of hundred closed-won deals at minimum and at least a year of history; a team closing a handful of deals a month should start with a rule-based score. - Use the data you have. "No company has perfect data. What really matters is how consistent the data is," Pete says. A model can learn how your team records pipeline, even when that record is imperfect. - Keep it simple at first. Start with five to seven signals. A model too complex to explain gets ignored. Keep manual inputs to two or three fields per deal and pull engagement signals (meetings, email activity, stakeholder count) from the CRM and communication tools. - Score each motion separately. A high-volume commercial motion, an enterprise motion with long cycles and a product-led motion behave differently. One pass line and one set of weights across all three repeats the stage-weight mistake. - Calibrate quarterly. Check which signals actually predicted outcomes and adjust weights. A signal that tracked wins last year may stop tracking them if buyer behavior shifts.
Common mistakes with pipeline scoring
Scoring deals once and never updating. A deal scored at 85 when it entered evaluation may be a 45 three weeks later if the champion went silent and a competitor entered. Scores must update as deal signals change, and the pipeline score should be tracked weekly so the trend shows. Letting reps self-score without validation. If reps can rate their own champion strength as "strong" without evidence, the model inherits their optimism. Require observable evidence for each signal. Champion strength means "champion presented business case to leadership." "My contact says they like us" does not count. Trusting old scores on stale deals. A deal untouched for 12 months keeps whatever score it had when it last moved. Flag stale deals first and keep them out of the passing value, whatever their last score says.What a pipeline score cannot tell you
A pipeline score measures the deals you can see, and part of every quarter comes from deals you cannot see yet. Pete splits a quarter into three sources: carry-over deals already in the pipeline on day one, deals created and closed inside the quarter, and deals pulled forward from future quarters. On ORM's data, about 20% of the pipeline carrying in-quarter close dates on day one usually closes in that quarter, varying by customer (how much day-one pipeline actually closes). On the $3,500,000 example above, that would be about $700,000. Scored coverage tells you how much of the visible pipeline is real; the forecast still has to model the rest.Frequently Asked Questions
What is pipeline scoring?
Pipeline scoring is the practice of scoring every open deal on evidence that it will close, then adding those scores up to judge the pipeline as a whole. The deal scores tell reps and managers where to spend time. The roll-up tells leaders how much of the pipeline is real and whether that real part is enough to cover the target.
What is a pipeline score?
A pipeline score is the share of open pipeline value sitting in deals that pass your deal scoring threshold, with stale deals kept out of the passing value. If $2,000,000 of a $3,500,000 pipeline passes, the pipeline score is 57%. Read it next to raw coverage: a pipeline at 3.5x with a score of 57% has 2.0x of coverage that is actually moving.
What is the pipeline scoring formula?
Pipeline score equals the value of open deals at or above the pass line, divided by total open pipeline value, times 100. Scored coverage equals that passing value divided by the remaining target. Each deal's score comes first: a weighted sum of engagement, process and qualification signals, or a probability from a predictive model. Stale deals never count as passing.
How does pipeline scoring differ from lead scoring?
Lead scoring ranks prospects before they enter the pipeline and asks whether a lead should become an opportunity. Pipeline scoring rates opportunities already in the pipeline and asks how likely each one is to close and how much of the total is real. Lead scoring is a top-of-funnel tool for marketing and SDRs. Pipeline scoring is a forecasting and coaching tool for sales leaders.
What signals should pipeline scores use?
Start with signals of movement: a recent change in stage, close date or amount, and the direction of that change. Add stakeholder count, champion strength, executive sponsorship, competitive position, time in stage against the median, budget confirmation and a defined next step. Weight each signal by how strongly it separated your past wins from your past losses, and score each sales motion separately.
How much can pipeline scoring improve forecast accuracy?
Scoring improves forecast accuracy because it finds at-risk deals earlier and stops stage probabilities from treating every deal at a stage the same. The size of the gain depends on your data and your sales motion. To measure it on your own pipeline, compare forecast accuracy in the quarters before and after scoring goes live, using the same definition of accuracy.
Put these metrics to work
ORM builds custom revenue forecast models that turn concepts like pipeline scoring into prescriptive action for your team.
Schedule a Demo