Lead Scorer

Sales Cycle Length: Measure the Real Delay, Not Just the Average

Calculate sales cycle length consistently, find the stage where deals stall, and improve buyer progress without gaming the pipeline metric.

By Miljan @ Lead Scorer 9 min read

A shorter sales cycle length is not automatically a healthier sales process. Close long opportunities as lost, create opportunities later, or move stage dates by hand and the average improves on paper while buyers move no faster. The useful question is: where does a qualified deal wait, and what evidence would let the buyer take the next step?

Editorial diagram separating sales cycle duration from time lost between buyer decisions

Define the clock before calculating it

A sales cycle describes the steps from first sales engagement to a buying decision. Its measured length depends on the start event you choose. Some teams start at first contact; others start when an opportunity is created. Both can be useful, but they answer different questions. If marketing qualifies a lead for weeks before sales opens an opportunity, the opportunity clock hides that waiting period.

For an opportunity-level metric, write a rule such as: start at opportunity creation; stop at closed won; report closed lost separately. Preserve the original creation date when stages change. Track first inquiry to first response as a separate measure, not an interchangeable version of the cycle. The distinction matters if you already monitor speed to lead.

Pipedrive's deal-duration documentation illustrates why the filter must be explicit: its duration report can show won, lost, or open deals, and its average reflects the records selected. A change in filters can change the number without changing a single buyer decision.

Calculate the mean, then show what it hides

For a cohort of closed-won opportunities, calculate the days from the agreed start event to close for each deal. Add those durations and divide by the number of wins. In a hypothetical cohort with durations of 20, 30, and 100 days, the mean is 50 days while the median is 30. Neither is wrong. The gap tells you that a single long deal affects the average substantially.

Put four numbers together whenever you report the metric:

  • Mean duration for a comparable closed-won cohort.
  • Median duration to expose the typical deal when outliers exist.
  • Number of deals so a small sample is visible.
  • Win rate so faster disqualification does not masquerade as faster selling.

Keep an open-deal view too, but call it current age, not completed cycle length. The age of an unfinished opportunity is still growing. Lost-deal duration is useful for diagnosing wasted effort, yet mixing it with won deals makes the headline metric hard to interpret.

Break the cycle into buyer commitments

An average tells you that a delay exists, not where to act. Record when a deal entered and left each stage. Then inspect time in stage, conversion, and the next buyer commitment. A stage should represent an observable change, not an optimistic salesperson label. This is also the foundation of a reliable sales dashboard.

StageEvidence to captureQuestion when time grows
DiscoveryProblem and affected workflow confirmed by the buyerWas the opportunity opened before a real problem was established?
EvaluationSuccess criteria and stakeholders identifiedIs a decision-maker or technical reviewer missing?
ProposalScope, buying steps, and a buyer-owned next dateDid the team send a proposal before agreeing the decision process?
ApprovalLegal, security, budget, and procurement owners knownWas a predictable review discovered too late?

Do not interpret every long stage as a seller failure. A buyer may have a real approval cycle that cannot be compressed. The operational failure is having no owner, no next decision, or no evidence explaining the wait. Review the buyer's own timeline before prescribing another follow-up, and keep that explanation with the opportunity record.

Compare like with like

Segment by motion before deciding whether the cycle changed. Inbound demo requests, outbound prospecting, existing-customer expansions, and enterprise procurement can have different start points and buyer paths. A shift toward larger accounts may make the blended average longer even if execution improves within each segment.

  1. Choose one reporting window and one start/stop definition.
  2. Separate new business from expansion, then split by deal size or customer segment if the sample supports it.
  3. Compare mean, median, win rate, and deal count within each group.
  4. Review the longest stage and a sample of the underlying records, not only the chart.
  5. Note stage-definition or CRM-policy changes before claiming a trend.

This discipline also improves sales forecasting. A forecast built on one blended cycle assumption can be misleading when the active pipeline contains several buying motions.

Run one intervention at a time

Once you find the slowest stage, choose a change tied to evidence. If discovery takes too long because account fit is unclear, tighten the entry rule and inspect the qualification record. If proposals stall because procurement appears late, ask about the approval path in discovery and record the owner. If deals wait without a next action, make the next buyer commitment and its date mandatory before advancing the stage.

Recheck the same cohort definition after the change. A useful improvement reduces unexplained waiting while preserving or improving win rate and buyer fit. If cycle length falls only because weak deals disappear from the report, record that honestly as better qualification, not faster decision-making. Lead Scorer's lead scoring guide explains how to separate fit and signals before expensive sales work begins.

A practical weekly review

Start with the deals that are older than comparable opportunities in the same stage. For each, confirm the last buyer action, next buyer commitment, decision owner, and reason for the delay. Advance only when new evidence exists; otherwise agree a recovery action, pause the deal, or close it with a reason. Then look at the cohort-level mean, median, win rate, and stage durations. The review turns a vague ambition to “shorten the cycle” into a specific decision about work.

If your first problem is choosing which accounts deserve that work, try Lead Scorer to prioritize prospects before they enter your active sales process. Keep buyer decisions and close dates in your CRM; use the evidence behind qualification to spend seller time deliberately.

Frequently asked questions

What is sales cycle length?

Sales cycle length is the elapsed time between a consistently defined start event and a closed deal. For opportunity reporting, a useful convention is opportunity creation to closed won, with lost deals reported separately. Record the convention before comparing teams or periods.

How do you calculate average sales cycle length?

For a defined cohort of closed-won opportunities, subtract each opportunity's start date from its close date, sum those durations, then divide by the number of wins. Report the median and sample size alongside the mean because a few unusually long deals can distort it.

Should open and lost deals be included?

Do not silently mix them into the closed-won average. Show lost-deal duration separately and treat open deals as current age or time in stage. An open opportunity has not finished its cycle, so a completed-cycle figure would be misleading.

What is the best way to shorten a B2B sales cycle?

First identify the stage and segment responsible for delay. Then fix a specific cause, such as missing qualification evidence, an unknown buying process, late legal review, or an unowned next step. Check win rate and deal quality alongside duration so a faster metric does not hide worse selling.

Keep reading