Sales pipeline

Sales forecasting grounded in CRM data

Build forecasts from pipeline hygiene: stages, amounts, and close dates teams agree on.

Reviewed August 2026 3 min read By the Vertex CRM team

Forecasts fail for one of two reasons: the underlying data is fiction, or the model applied to it is wrong. Almost always it is the first. No weighting scheme rescues a pipeline where close dates are decorative.

Pick a model that matches your deal count

The three common approaches suit different shapes of business. Running the wrong one produces confident nonsense.

Which forecasting model fits your deal shapeDeal-by-deal judgementEach deal reviewed on its merits.The only honest model when one dealdecides the quarter.Category forecastCommit / best case / pipeline, setby the owner and challenged weekly.Judgement, lightly weightedSmall enough to inspectindividually; use stage probabilityonly as a sanity check.Stage-weightedHistorical conversion by stageapplied at volume. Needs enoughdeals to average out.Larger dealsSmaller dealsFew deals per quarterMany deals per quarter
Deal count and deal size decide the model. High-volume, low-value pipelines average out and reward statistical weighting; low-volume, high-value pipelines do not—one deal moves the quarter, so judgement beats arithmetic.

Commit, best case, pipeline

Three categories carry more information than a single weighted number, because each one asks a different question:

  • Commit — the owner will personally answer for this landing this period. Missing commit is a serious conversation, and it should be.
  • Best case — realistic upside if things go well. Not a wish list; every entry needs a reason it could close.
  • Pipeline — everything else that is live. This is where coverage lives, not where the forecast lives.

The useful metric is not the forecast itself but the gap between commit and actual, tracked by owner over several quarters. Consistent optimism is a coaching problem; consistent pessimism is a sandbagging problem. Both are invisible without the history.

The weekly cadence that keeps it honest

A forecast cadence across a quarterwk 0wk 2wk 4wk 6wk 8wk 10wk 12Weekly pipeline hygieneMonthly category reviewMid-quarter coverage checkQuarter-end commit lockAccuracy retrospectivepost-close
Cadence matters more than sophistication. Weekly hygiene, monthly category review, and a quarter-end retrospective on forecast accuracy produce better numbers than any model applied to stale data.

Coverage, honestly

Coverage ratios—pipeline value divided by target—are only meaningful next to your own historical conversion. A team converting one in five needs materially more coverage than one converting one in three, and the multiplier every vendor quotes is somebody else’s number. Compute yours from closed history before you quote a rule of thumb.

Model the arithmetic on the pipeline management page, where the interactive funnel shows how a change at the top compares with a change at the bottom.

What Vertex CRM provides

Vertex CRM supplies structured opportunities, stage history, and reporting surfaces to slice by owner and period. It deliberately does not impose a forecast methodology—commit categories and weighting belong to your leadership model, not your vendor’s. Read CRM reporting and analytics for the report layer and opportunity management for the field-level hygiene this depends on.

Frequently asked questions

Why do weekly pipeline reviews matter?
Short reviews that update close dates and disqualify dead deals improve forecast accuracy faster than complex models built on stale data.
What is commit vs best case?
Commit is what leadership believes will close; best case includes upside that is not yet proven.

Ready when your team is

Bring your stages, owners, and messy spreadsheet—we’ll map it into a pipeline your leadership can defend.