Every forecast call runs on about twenty words. In a lot of the teams I’ve worked with, nobody ever wrote down what those words mean, so “commit” means one thing to the rep who said it, another to the manager who rolled it up, and a third to the CFO who planned hiring around it.
These are the definitions we use. Each one stands on its own, so you can lift it straight into a forecast process document. Where a CRM uses its own label, we say so.
What is the difference between commit, best case and pipeline?
Commit, best case and pipeline are forecast categories that sort open deals by how confident the seller is that they close this period. Commit is the number you will be held to. Best case is commit plus realistic upside. Pipeline is qualified opportunity not expected this period. HubSpot and Salesforce both use these three names, plus a closed category and an excluded one.
| Category | What it means | Evidence bar we would expect | How it is used |
|---|---|---|---|
| Commit | Will close this period; the rep stakes their number on it | Buyer-confirmed date, signer engaged, paper or procurement in motion | The number leadership plans from |
| Best case | Could close this period if it breaks right | Real buyer engagement and a plausible path to signature in time | Upside the team works to pull in |
| Pipeline | Qualified, not expected this period | Confirmed need and an active buyer conversation | Coverage for future periods |
| Closed won | Signed and booked | A signed contract | Actuals |
| Not forecasted / Omitted | Excluded on purpose | None | Closed lost, paused, duplicates |
Platform labels differ slightly. HubSpot’s forecast tool uses Pipeline, Best case, Commit and Closed won, plus Not forecasted, and can map deal stages to categories automatically. Salesforce’s standard categories are Pipeline, Best Case, Commit, Closed and Omitted, and changing an opportunity’s stage changes its category by default. Pipedrive has no categories in this sense: its forecasting runs on stage and deal probability. Wherever stage drives the category automatically, the category stops carrying anyone’s judgment, which is the point of having one. We go further into that in why your HubSpot forecast is always wrong.
Forecast categories, defined
A forecast category is a label on each open deal that says how confident the seller is that it closes in the period, independent of its stage. Stage records where the deal is in your process; category records the seller’s call. The categories below are the standard set. Most teams need the first three, a closed bucket and an excluded bucket, and no more.
Commit
Also: the call, commit forecastCommit is the forecast category for deals a rep or manager expects to close in the current period and is willing to be held to. The commit total is the number leadership plans from, so it needs a written evidence bar, typically a buyer-confirmed close date, an engaged signer and contract or procurement steps under way. A commit that misses regularly is a definition problem before it is a rep problem.
Best case
Also: upsideBest case is the forecast category for deals that could close this period if the remaining steps go the seller’s way. The best-case total is commit plus that upside. Best case is where a forecast call earns its keep: these are the deals a manager can help pull in, and the gap between commit and best case shows how much of the period still depends on breaks.
Pipeline (category)
Not the same as “the pipeline”As a forecast category, pipeline holds qualified open deals that are not expected to close in the current period. They count toward coverage for future periods, not toward this period’s call. The word also means all open opportunities in general, so say “pipeline category” when you mean the bucket. Otherwise the two meanings blur and the forecast call argues about vocabulary.
Closed won
Also: closed, bookingsClosed won is a deal the customer has signed and the business has booked. In a forecast it is actuals, not a prediction. Two details make it messier than it sounds: the close date recorded in the CRM can differ from the signature date when reps update late, and sales often counts bookings while finance counts recognized revenue. Agree which one the forecast is measured against.
Not forecasted / Omitted
HubSpot: Not forecasted. Salesforce: OmittedNot forecasted, in HubSpot, or Omitted, in Salesforce, is the category for deals deliberately excluded from the forecast: closed lost, paused, duplicated or opportunities that should never have been created. It is a legitimate bucket. When it grows large, read it as a sign that the CRM is carrying deals nobody believes in instead of closing them out.
Most likely
Also: expected, single-number forecastMost likely is a single forecast number between commit and best case, used by teams that ask reps for one expected outcome instead of a floor and a ceiling. It is the most honest number a rep can give and the hardest to hold anyone to. Many teams track it alongside commit, as a read on where the rep thinks the period lands, rather than instead of it.
Pipeline math: weighted pipeline, coverage and accuracy
Pipeline math turns a list of deals into a number. Weighted pipeline multiplies each deal by a probability, coverage compares open pipeline with the target, and accuracy and bias grade the forecast after the period closes. Every one of these depends on inputs a person typed, which is why the definitions matter less than the hygiene underneath them.
Weighted pipeline
Also: weighted forecast, weighted amountWeighted pipeline is the sum of each open deal’s amount multiplied by its probability of closing. A $100,000 deal at 40% contributes $40,000. It is only as good as its probabilities: HubSpot’s default pipeline assigns 20% to 90% by stage for every company. Calibrate the probabilities to your own stage-to-close win rates, or the weighted number is a weighted guess.
Deal stage probability
Also: win probability, stage probabilityDeal stage probability is the win probability a CRM assigns to every deal in a given pipeline stage. In HubSpot it sets each deal’s Deal probability automatically when the deal changes stage; in Pipedrive every stage starts at 100% until someone changes it. It treats a deal that entered the stage yesterday the same as one that entered eight months ago.
Pipeline coverage
Also: coverage ratioPipeline coverage is open pipeline expected to close in a period divided by the target for that period. Three times coverage is a common rule of thumb, not a law. The coverage you need is roughly one divided by your win rate, plus a buffer for deals that slip: a team that wins one deal in four needs more than 3x, a team that wins one in two needs less.
Forecast accuracy
Formula: 1 minus |actual minus forecast| divided by actualForecast accuracy is how close a forecast landed to what closed, usually 1 minus the absolute miss divided by actual. A $1.0M forecast against $0.92M closed is 91.3% accurate. Xactly’s 2024 benchmark found only 20% of sales organizations forecast within 5% of actuals. Measure from a saved snapshot, such as the first-month call, or the number means nothing.
Forecast bias
Formula: (forecast minus actual) divided by actualForecast bias is the signed version of forecast error: forecast minus actual, divided by actual. Positive bias means you called high; negative bias means you called low. For diagnosis, bias matters more than accuracy, because a consistent lean points to a cause, such as stale pipeline when you call high or sandbagging when you call low.
Slippage
Also: deal slippage, close date pushSlippage is a deal forecast to close in one period that moves to a later one, usually because a rep changed the close date. Some slippage is normal. Repeated slippage on the same deal is one of the most reliable signs that a forecast is softer than it looks, which is why some teams keep a push count on every deal and review it in the forecast call.
What do sandbagging and happy ears mean in sales?
Sandbagging means deliberately forecasting less than you expect to close; happy ears means hearing buying signals that are not there and forecasting too much. They are the same error pointed in opposite directions: the rep managing the forecast instead of reporting it. Both leave fingerprints in CRM data, alongside two related patterns, single-threaded deals and stale pipeline.
Sandbagging
Forecasting low on purposeSandbagging is deliberately forecasting less than you expect to close, so you can beat the number or protect against a miss. It looks safe to the rep and costs the business, because leadership plans hiring and spend from a number the rep knows is low. In CRM data it shows up as deals created and closed in the same period and commits that run consistently under actuals.
Happy ears
Hearing a yes that was not saidHappy ears is hearing buying signals that are not there: treating a friendly champion, a good demo or a pricing request as commitment. It is a leading cause of over-forecasting. In CRM data it looks like late-stage deals with no buyer activity in weeks, close dates pushed again and again, and single-threaded deals sitting in commit.
Single-threaded deal
One contact, one point of failureA single-threaded deal is an opportunity where the seller has a relationship with one person at the buying company. If that person leaves, goes quiet or loses the internal argument, the deal stalls with no fallback. In the CRM it shows up as a deal with one associated contact, and it deserves less confidence than its stage implies.
Stale pipeline
Open, but not movingStale pipeline is open deals that are no longer progressing but have not been closed out: no recent activity, no next step, often a close date already in the past. It inflates coverage and weighted pipeline at the same time, which makes it the quietest way a forecast gets overstated. Close it lost, or re-date it with a written reason.
Why reps fall into these patterns, and what managers can do about it, is the subject of why sales reps are bad at forecasting. The incentive side is in why good reps enter bad data.
The forecast call and the roll-up
The forecast call is where the categories get decided and the roll-up is how the decisions travel upward. Both work when they inspect buyer evidence and leave a record of who changed what. Both fail quietly when they run on adjectives and silent overrides, because nobody can learn from a miss they cannot trace back to a decision.
Forecast call
Also: forecast review, deal reviewA forecast call is the recurring meeting, usually weekly, where reps and managers review the deals behind the number and agree the commit and best case for the period. A good forecast call inspects buyer evidence deal by deal: what the buyer did, who signs, what date they gave. A weak one asks each rep how they feel and writes down the answer.
Forecast roll-up
Also: forecast submissionA forecast roll-up is the chain of submissions behind the company number: reps submit, managers adjust and submit for their teams, and leaders submit for the business, each layer adding judgment. The roll-up is only as honest as its least honest layer, so record who changed what at each level and why. Without that record, nobody can learn from a miss.
Manager override
Also: judgment, adjustmentA manager override is an adjustment a manager makes on top of a rep’s call, based on what they know about the deal or the rep’s track record. Overrides are healthy when they are visible and explained, because reps learn from them. Silent overrides teach reps that their call does not matter, and the next call gets less careful.
Forecast snapshot
The saved callA forecast snapshot is a saved copy of the forecast at a fixed point in the period, such as the end of week two. CRMs recalculate the live forecast as deals move, so without snapshots you cannot measure accuracy later. Snapshot by rep and in total, keep four or more periods, and grade every period against the same snapshot point.
To grade the forecast these terms produce, use the formulas in our sales forecast accuracy guide. To check whether the deals behind the categories are real, the AeolusGTM CRM Diagnostic runs a read-only scan of HubSpot, Salesforce or Pipedrive and shows open deals missing the numbers a forecast needs, stalled deals and close dates that keep sliding. HubSpot teams can start with the free HubSpot audit.
Write the definitions down this week, and put them at the top of the next forecast call. The first argument they settle will pay for the hour.