Section 1
The five rates that explain your revenue
Every service pipeline reduces to five transitions. Lead-to-contact: of new leads, how many did you actually reach? Contact-to-booked: of those reached, how many scheduled a call? Booked-to-held: how many showed up? Held-to-proposal: how many calls earned a proposal? Proposal-to-won: how many proposals signed? Multiply the five and you get your true lead-to-client rate, typically somewhere between 5% and 20% for service firms, a range wide enough to be the difference between struggle and scale on identical marketing spend. The arithmetic has a useful property: improving any single stage lifts revenue proportionally, but stages compound, so a pipeline weak in two places is far worse than twice as bad. Measure first. Most owners, seeing their five numbers for the first time, immediately know what their last six confusing months were actually about. To see how this connects to the wider system, read [From Lead to Booked Call: Designing the Follow-Up System That Closes](/blog/from-lead-to-booked-call-follow-up-system).
Section 2
Benchmarks and red flags by stage
The table below shows working benchmarks drawn from the service businesses we audit and install LeverageOS into, alongside the red-flag thresholds that should trigger investigation. Treat benchmarks as orientation, not gospel, referral-heavy pipelines run hotter than cold-traffic pipelines at every stage. What matters more than any single number is trend and comparison: your stages against each other, this quarter against last. Note how the early stages are mostly speed and systems problems (HBR's lead-response research explains lead-to-contact almost entirely), while later stages are mostly offer, qualification, and sales-craft problems. The diagnosis differs by stage, which is exactly why a single blended conversion number hides more than it reveals. Investigate causes stage by stage before changing anything downstream.
Section 3
Find the constraint, ignore the rest (for now)
Once the five numbers exist, resist the urge to improve everything. Theory of Constraints logic applies: one stage is the bottleneck, and effort anywhere else is largely wasted until it moves. Find it by asking where the gap between your number and the healthy range costs the most clients in absolute terms, a show rate of 55% usually outranks a proposal rate of 28%, because it starves every stage after it. Then fix that stage with its known toolkit (reminder sequences for show rate, follow-up sequences for proposal rate, response systems for contact rate), measure for a month, and re-locate the constraint, because it moves. Deming's warning, that a bad system beats a good person every time, is the right lens for what the numbers show you: a low stage rate is almost never a people problem, and treating it as one burns morale while the system keeps losing. For a deeper look at this, see [The Lead Generation Stack for Service Businesses: What You Need at Each Stage](/blog/lead-generation-stack-service-businesses).
Section 4
Instrumenting this without a data team
You need nothing exotic: a CRM whose stages match the five transitions, a discipline of moving deals on observable events, and a monthly thirty-minute review. Count cohorts, not snapshots, of the forty leads that arrived in April, how many reached each stage, so slow-moving deals do not blur the picture. Most lightweight CRMs report this natively once stages are clean; a spreadsheet works at low volume. Salesforce's State of Sales research notes high-performing teams are far heavier users of data in decision-making, but for a small firm the bar is wonderfully low: five numbers, monthly, honestly counted, puts you ahead of most of your market. HubSpot's compiled statistics tell the same story, teams that inspect their funnel outperform teams that feel it. This dashboard is the first thing a LeverageOS install switches on, and a strategy call can usually estimate your five numbers from a single conversation. A useful companion to this piece is [The Psychology of Website Conversion: Why Buyers Say Yes (or Quietly Leave)](/blog/psychology-of-website-conversion).