A £16,000 Google Ads quarter can produce 80 “conversions”, a £200 cost per lead and almost no usable evidence about sales performance. If only 12 enquiries become qualified opportunities and four become clients, the headline dashboard has measured the cheapest event, not the commercial outcome.
Useful B2B sales cycle metrics do three things: start and stop at documented events, retain the cohort long enough to mature, and connect elapsed time to qualified pipeline or booked revenue. Everything else is either a diagnostic or decoration.
What B2B sales cycle metrics should tell you
Before tracking, give each metric a measurement contract: unit, start and stop events, denominator, cohort, clock and exclusions. A win-rate trend is meaningless if the denominator changes.
A useful metric also needs an explicit observation window, a decision owner and resistance to gaming.
Context still matters. Procurement, deal value and wider buying conditions can alter the baseline; our analysis of the market forces lengthening B2B sales cycles covers that problem. Measurement has a narrower job: establish what changed, where and by how much.
The B2B sales cycle metrics worth tracking
No single number explains a pipeline. A compact dashboard needs elapsed-time measures, conversion measures and one throughput view.
1. Sales-cycle length as a distribution
Use a fixed clock:
cycle days = signed contract date − qualified opportunity date
Use calendar days consistently. Do not start when a salesperson gets around to creating the CRM record, and do not stop at a verbal yes.
For closed-won deals, publish the mean, median, 75th percentile and sample size. The median describes the typical win; the mean retains unusually slow deals; the 75th percentile exposes the long tail. Report lost-deal duration and current open-deal age separately. Combining all three creates a number nobody can interpret.
External comparisons only help when the start point and commercial segment match. Use benchmarks split by deal size and sector as a sense check, not a target handed to every salesperson.
2. Enquiry-to-qualification time and rate
The website and paid media affect the period before an opportunity exists, so keep a second clock:
qualification time = qualified opportunity date − valid enquiry date
Also calculate:
qualified-opportunity rate = qualified opportunities ÷ valid enquiries
“Valid” should exclude spam, tests and duplicates, using a documented rule. Report the median time and rate by source, campaign and landing page. If inbound routing is part of the problem, add time to first human response; an automated receipt does not count.
This pair separates slow follow-up from weak traffic. Forty enquiries producing 20 suitable opportunities beats 200 producing ten, whatever the lead-volume slide says.
3. Cohort win rate
The defensible count formula is:
win rate = closed won ÷ (closed won + closed lost)
Open opportunities do not belong in the denominator. Group opportunities by the period in which they qualified, then wait until the observation window is mature. A report of deals closing this month answers “what closed now?” It does not tell you how March’s paid-search cohort performed.
Track value win rate alongside count win rate when deal sizes vary. Use one amount basis—such as first-year contract value—and one currency. A 40% count win rate can conceal repeated losses on the largest deals.
4. Open-deal age, stage dwell and slippage
For every open opportunity, calculate today − qualified opportunity date. Compare it with the historical 75th percentile for similar deals. The percentage beyond that threshold is an ageing rate and an investigation queue, not proof that every old deal is dead.
Stage dwell adds location: how long the deal has spent at its current recorded checkpoint. This assumes those checkpoints are already defined consistently; the metric should not smuggle in a new stage model.
Close-date slippage shows repeated optimism:
slippage rate = deals due at period start whose expected close moved later ÷ deals due at period start
Report the count, value and median days added. Current CRM fields cannot reconstruct this after the old date is overwritten, so slippage requires field history or snapshots.
5. Deal velocity
Cycle length alone misses throughput. Use:
deal velocity = open qualified opportunities × average deal value × matched win rate ÷ mean cycle days
The result is a directional £-per-day operating index, not booked revenue or a cash forecast. Every input must come from the same sales motion and value band. The four-lever deal-velocity model explains how volume, value, win rate and time interact without pretending the output is guaranteed revenue.
6. Cost per qualified opportunity and won client
For paid channels, retain two blunt calculations:
cost per qualified opportunity = channel spend ÷ qualified opportunities
cost per won client = channel spend ÷ wins from the matured acquisition cohort
Segment them by original source and campaign, but treat attribution as allocation rather than proof of causation. These figures reveal whether cheaper clicks and forms become commercially useful pipeline.
A worked example: £200 per lead becomes £4,000 per win
A B2B software company spends £16,000 on Google Ads in one quarter. Its form conversion is recorded 80 times. After spam, duplicates and sales follow-up, 32 are valid enquiries and 12 become qualified opportunities.
After allowing the cohort 180 days to mature, all 12 are resolved: four won and eight lost. The four wins produce £96,000 in first-year bookings and take 42, 51, 64 and 151 days from qualification to signature.
The calculations are:
- Reported cost per form: £16,000 ÷ 80 = £200
- Cost per valid enquiry: £16,000 ÷ 32 = £500
- Qualified-opportunity rate: 12 ÷ 32 = 37.5%
- Cost per qualified opportunity: £16,000 ÷ 12 = £1,333
- Win rate: 4 ÷ 12 = 33.3%
- Cost per won client: £16,000 ÷ 4 = £4,000
- Mean won cycle: (42 + 51 + 64 + 151) ÷ 4 = 77 days
- Median won cycle: (51 + 64) ÷ 2 = 57.5 days
The £200 figure prices an action before quality is known. The £4,000 figure arrives later but can be compared with gross profit and cash economics. The median describes a typical win; the 77-day mean retains the long tail.
Bookings divided by spend equals 6.0, but that is not automatically 6.0x incremental ROAS or recognised revenue. The CRM establishes an association; another method must establish credit. The same standard applies when assessing suppliers: case studies with named commercial outcomes are more informative than screenshots of impressions or clicks.
How to instrument sales cycle metrics
The minimum viable setup is not a bigger dashboard. It is reliable event history joined to a stable opportunity ID.
Capture these fields:
- Opportunity and account IDs, with a clear duplicate and merge rule
- Valid enquiry, first human response, qualification and actual close timestamps
- Outcome, final amount, currency, value basis and new-business or expansion flag
- Original source, campaign, landing URL and advertising click identifier
- Append-only stage-entry and exit history, including re-entry
- Expected close-date and amount-change history, or a daily opportunity snapshot
Store raw timestamps consistently and calculate calendar days in the reporting layer. Keep original-source data immutable, with a separate latest-touch field if needed.
The website-to-CRM handoff deserves a physical test. Submit every important form, confirm that the record arrives once, check that source fields survive, and verify that the opportunity outcome joins back to the originating campaign. When a site changes, the redesign measurement and launch checklist helps preserve baselines and tracking continuity rather than discovering the break three months later.
Closed-loop source data is the measurement spine of a source-to-revenue growth model: web analytics provides touch evidence, the CRM records commercial progression, and finance confirms booked or collected value. Advertising platforms should receive qualified and won outcomes only within the company’s consent and privacy setup.
Finally, publish a one-page metric dictionary covering each formula, owner, source field, exclusion, cadence and definition version. Annotate any CRM migration or clock change; do not splice incompatible definitions into one trend line.
Review cadence: weekly signals, monthly decisions
Different metrics mature at different speeds.
| Cadence | Review | Decision |
|---|---|---|
| Continuous or daily | Missing timestamps, duplicate records, broken form routing, impossible date sequences | Repair the data before it contaminates reports |
| Weekly | Human-response time, open age, current dwell and close-date slippage | Investigate specific delays and routing failures |
| Monthly | Matured-cohort win rate, cycle distribution, cost per opportunity, cost per win and velocity | Adjust budget, capacity or experiments by segment |
| Quarterly | Metric definitions, cohort maturity, source mapping, thresholds and finance reconciliation | Reset baselines and approve larger changes |
Use alerts for exceptions, not instant verdicts. A practical trigger might be a median or 75th percentile worsening by more than 15% across two mature monthly cohorts. One delayed enterprise deal should not cause a website rewrite.
Always show the sample size and unresolved count. If a segment has fewer than 20 decided opportunities, extend the rolling window and show the raw numbers. “Win rate doubled” means little when it moved from one win in four decisions to two.
The reporting lag must match the cycle. If the 75th percentile is 150 days, do not judge August’s campaign on October’s revenue. Use qualification rate and cost per qualified opportunity as earlier signals, then replace provisional views with won-client economics when the cohort matures.
Vanity metrics to demote
A metric is not vain because it sits near the top of the funnel. It becomes vanity when presented as commercial progress without quality, time or revenue attached.
| Flattering headline | Better operating view |
|---|---|
| Sessions, impressions or click-through rate | Qualified opportunities and wins per mature source cohort |
| Form fills or MQL volume | Valid-enquiry rate, qualification rate and cost per qualified opportunity |
| Total pipeline value | Age distribution, slippage, matched win rate and deal velocity |
| Average sales cycle alone | Mean, median, 75th percentile and sample size, with won, lost and open views separated |
| Calls, emails or meetings booked | First human response, meetings held and downstream cohort conversion |
| Platform-reported ROAS | Deduplicated CRM bookings, gross profit and cost per win after cohort maturity |
Keep diagnostic measures where they help somebody act. Click-through rate can identify a weak advert. Form abandonment can expose a broken page. Neither is a revenue result.
Every dashboard tile needs a denominator, observation window, segment and decision owner. Without them, it belongs in an investigation report rather than the board pack.
FAQ
What is the most important B2B sales cycle metric?
Median qualified-opportunity-to-signature time is the clearest typical measure, but it needs the 75th percentile and sample size beside it. Pair it with qualified-opportunity rate and matured-cohort win rate so speed is not mistaken for quality.
When should the sales-cycle clock start and stop?
Start when an opportunity meets the documented qualification rule. Stop at signed contract for wins and the actual decision timestamp for losses. Record first enquiry separately, use calendar days consistently and never reset the start when a deal reopens.
Should open and lost opportunities be included in average cycle length?
Do not blend them into the won-cycle average. Report won duration, lost duration and open age separately. Treating open deals as zero days—or ignoring their existence—makes recent cohorts look artificially fast.
Does a shorter sales cycle always mean better performance?
No. The company may have shifted towards smaller deals or begun rejecting poor-fit opportunities faster. Read cycle length with deal value, win rate, qualified volume and gross profit.
How often should sales-cycle metrics be reviewed?
Review data failures and ageing exceptions weekly. Review mature cohorts monthly, or quarterly when volume is low. Reconcile with finance and revisit definitions quarterly. Budget decisions should follow cohort maturity, not a convenient calendar month.
Summary
- Define every metric’s events, denominator, cohort, clock and exclusions before building the dashboard.
- Track cycle distribution, qualification, win rate, ageing, slippage, velocity and paid-channel unit economics.
- Join website source data to CRM outcomes and preserve timestamp and close-date history.
- Review exceptions weekly, mature performance monthly and definitions quarterly.
- Treat traffic, form fills, activity and platform ROAS as diagnostics until they connect to commercial outcomes.
Actualyse builds websites and campaigns designed for long, committee-driven B2B sales cycles. Book a call to talk through where yours stands.
