“We should be closing about 30% of our pipeline” is a statement built on a number that came from somewhere — but often no one can say exactly where. It might be an industry benchmark, a number a previous manager used, or a figure that felt reasonable at some point. What it rarely is: a rate calculated from your own actual sales data.

Forecasts built on assumed close rates produce assumed revenue numbers. When the quarter ends and the miss arrives, you trace it back to the gap between what you assumed would close and what your pipeline actually converts at.

Why Using Assumptions to Forecast Is Expensive

Industry benchmarks for close rates are averages across a wide range of companies, deal types, and sales motions. Your sales motion — your specific ICP, your deal structure, your typical sales cycle, your team’s approach — may perform dramatically differently from the average.

A company selling a $5,000/year product to SMBs through a transactional inside sales process has a different close rate than a company selling a $200,000 enterprise deal through a 6-month consultative process. Using one benchmark for both produces meaningless forecasts for both.

The cost of assumption-based forecasting compounds. When you consistently overestimate your close rate, you over-hire to support a revenue number that never materializes. You miss quota and trace the miss to a pipeline coverage problem — when the real problem was a forecasting model built on wrong inputs. When you consistently underestimate it, you constrain investment and miss growth opportunities.

What your own CRM historical data tells you that assumptions can’t: the actual stage at which most deals die, the specific deal sources that convert at the highest rates, the rep-level variation in close rates, and the deal size ranges that close reliably versus the ones that consistently slip.

Building Your Historical Close Rate Dataset

What to Pull from CRM

Start with all deals closed — both won and lost — over the last 12 to 24 months. The longer the period, the more statistically reliable your rates will be, but 12 months is usually sufficient for most teams to see meaningful patterns.

For each deal, you want: the date the deal was created, the date the deal was closed (won or lost), the stage at which the deal was created (or the stage at which it entered the pipeline if that’s different from the creation stage), the stage at which the deal was lost (for lost deals), the deal value, the deal source (how this lead originated), the rep who owned the deal, and the deal size segment if you track that field.

Entry date to each stage and exit date allows you to calculate stage-to-stage conversion rates and the time each deal spends at each stage — information that tells you both where deals die and how long it takes.

Segmenting Close Rates That Actually Matter

A single overall close rate — total deals won divided by total deals entered — tells you almost nothing useful for forecasting. You need the segmented view.

By pipeline stage: Stage-to-close rates tell you what percentage of deals that reached a given stage eventually closed as won. This is the data that feeds your weighted pipeline forecasting model. A deal that has reached your proposal stage might close at 48%. A deal still in discovery might close at only 22%.

By deal source: Inbound leads, outbound prospecting, referrals, and event-sourced leads often convert at radically different rates. If your referral close rate is significantly higher than your outbound close rate, your forecast should apply different rates to deals from each source.

By deal size: Small deals, mid-market deals, and enterprise deals frequently have different close rates and different sales cycle lengths. Combining them in a single close rate produces a blended average that may not accurately represent either segment.

By rep: Individual rep close rates are valuable for coaching, and they also affect your forecast accuracy. If you’re forecasting based on a team average but half your team is significantly below average, your forecast will be overoptimistic.

By industry or customer segment: If you sell across multiple industries and your deal volume supports it, segment-specific close rates help you apply the right probability to deals from each vertical.

Calculating and Applying Stage-to-Close Rates

The formula for a stage-to-close rate is simple: take the number of deals that reached a given stage and eventually closed as won, divide by the total number of deals that reached that stage, and express as a percentage.

For example: if 80 deals reached your “Proposal Sent” stage over the past 12 months, and 38 of them eventually closed as won, your stage-to-close rate from “Proposal Sent” is 47.5%.

To apply this to forecasting: look at all deals currently sitting at “Proposal Sent” in your pipeline. Sum their total values. Multiply that sum by 47.5%. The result is your expected revenue from that stage of the pipeline.

Sample size matters significantly. A stage-to-close rate based on five deals is not reliable. Before you trust a rate, you want a meaningful sample — at minimum 15 to 20 deals per stage per time period you’re calculating for. If you don’t have that yet, note that your rates are estimates and weight them accordingly in your forecast.

Close Rate Dimensions Reference

Close Rate Dimension	What to Calculate	Minimum Sample Needed	How to Apply	How to Interpret
Stage-to-close rate	% of deals at each stage that closed won	15-20 deals per stage	Weight each open deal by its stage rate	Low rates at early stages are normal; low rates at late stages are a problem
By deal source	% close rate per lead source (inbound/outbound/referral)	20+ deals per source	Apply source-specific rates to current pipeline by source	High variation between sources reveals where to focus lead gen
By deal size	% close rate for small/mid/large deals	15+ deals per size band	Segment your pipeline into size bands, apply each rate	Large deals often have lower rates — don’t assume they’ll save the quarter
By rep	Individual rep close rate vs team average	All reps’ recent deals	Adjust forecast where certain reps carry large pipeline shares	Significant variation signals coaching opportunity
By industry segment	% close rate per vertical	15+ deals per vertical	Apply vertical rates to deals in that segment	Some verticals may not be worth continued investment
Stage-to-stage conversion	% of deals advancing from one stage to the next	15+ deals per transition	Identify which stage transition is the primary bottleneck	This reveals process problems, not just rep performance

Adjusting Forecasts Based on What Historical Data Reveals

Close rate patterns are often seasonal. If your business has consistently lower close rates in Q1 (because buyers are in planning mode and not yet ready to commit), applying an annual average rate to your Q1 pipeline will produce an over-optimistic forecast. Segment your historical data by quarter to see whether seasonal patterns exist.

Deal age affects close probability in ways that historical data makes visible. A deal that has been open for longer than your average sales cycle is not more likely to close — it is typically less likely, because deals that take longer than average often indicate qualification problems, internal obstacles on the buyer’s side, or stalled momentum. Your historical data will confirm whether this is true for your specific pipeline.

When you change something significant in your sales process — bringing on new reps, launching a new product, entering a new market — your historical close rates may not apply to the new situation. Be explicit about these adjustments in your forecast. Note that the rate for new-market pipeline is an estimate, not a historical calculation, until you accumulate enough data to calculate it empirically.

Building the Forecast

With your segmented close rates calculated, building the forecast becomes a structured process. Pull your current open pipeline. Segment it by the dimensions your close rate data supports — stage, source, size, rep. Apply the relevant close rate to each segment. Sum the expected values across all segments.

Layer in rep commit on top of the data-driven calculation. Ask each rep which specific deals they believe will close this period. Compare their commit to what the stage-to-close rate model predicts for their pipeline. Where there’s significant disagreement, dig into the reasons — either the rep knows something about the deals that justifies optimism, or the rep is overstating their commit.

Apply management judgment as a final layer. When you know a specific large deal has procurement complications that don’t show up in the stage data, or a key deal is tracking ahead of the historical average because of an unusually engaged champion, adjust accordingly and document why.

FAQ

How much historical data do we need before close rates are reliable? For a stage-to-close rate to be meaningful, you want at least 15 to 20 deals that have passed through each stage in the period you’re measuring. For a team doing 10 deals per month, 12 months of data gives you a reasonable sample for most stages. For teams with lower deal volume, you may need 18 to 24 months of data to build reliable rates at every stage.

What if our close rates are very low — does forecasting still work? Forecasting works at any close rate as long as the rates are consistent. Low close rates simply produce a more conservative weighted forecast — which is more accurate than an optimistic one built on assumed rates. The more important question is why your close rates are low: is it a pipeline quality problem (too many unqualified deals), a sales process problem (losing deals late), or a product-market fit problem (losing to “no decision” frequently)?

How do we handle a large deal that doesn’t fit the historical pattern? Large deals that fall significantly outside your normal deal range should be modeled separately from your standard pipeline close rates. Apply the stage-to-close rate as a baseline, then adjust based on what you specifically know about the deal: engagement level of the key stakeholders, presence of a clear champion, competitive situation, and buyer’s timeline. Track these adjustments explicitly in your forecast model.

Should we share close rate data with the sales team? Yes — with context. Individual close rates should be shared in private, one-on-one coaching conversations, not announced publicly. Team-level stage conversion data is useful for everyone to understand because it shows the team where deals typically die and focuses improvement efforts. When reps understand that their “Discovery to Proposal” conversion rate is lower than the team average, it becomes a specific, actionable coaching focus rather than a general performance critique.