One win rate tells you almost nothing
Ask most teams how their sales funnel is doing and you'll get a single number: "we close about twenty percent." It's a comforting number and a nearly useless one, because it averages away the only thing worth knowing — where the other eighty percent went. Deals don't evaporate evenly across your process. They pour out of one or two specific stages, and a blended win rate is precisely the statistic designed to hide which ones. Stage-by-stage conversion rate — the percentage of deals that advance from each stage to the next — is the measurement that turns "we lose a lot of deals" into "we lose them right here," which is the only version you can actually do something about.
The shift in thinking is from the funnel as a single tube to the funnel as a series of gates. Each pipeline stage has its own pass-through rate, and your overall win rate is just those rates multiplied together. That framing matters because it tells you a small fix at the leakiest gate beats a heroic effort spread evenly everywhere — and it tells you exactly which gate to look at first.
How to measure it, stage by stage
The calculation is simple; the discipline is in measuring it consistently. For each adjacent pair of stages, conversion rate is the share of deals that entered the earlier stage and made it to the next one:
- Lead to qualified: of the leads that entered the funnel, how many became real, qualified opportunities?
- Qualified to demo/proposal: of qualified deals, how many reached a serious evaluation?
- Proposal to negotiation: how many advanced from "here's what it costs" to actively working terms?
- Negotiation to closed-won: of the deals at the table, how many signed?
Run those numbers and you get a conversion rate per gate instead of one average for the whole journey. Two cautions make the numbers trustworthy. First, the funnel is only as honest as your pipeline hygiene — if dead deals linger in stages and stages mean different things to different reps, your conversion rates measure your data quality, not your selling. Second, give deals time to flow through before you judge a cohort, or you'll mistake deals still in flight for deals that leaked.
Read the pattern, because each leak means something different
Where the funnel leaks tells you what is broken, and the diagnosis is different at every gate. The shape of the leak is the message.
- Leaking at lead-to-qualified? You have a top-of-funnel quality or qualification problem. Either the leads coming in are poor fit — a targeting and ICP issue — or your reps are letting weak deals into the pipeline that were never going to make it. A low rate here isn't always bad: ruthless early qualification should disqualify a lot, and that's healthier than carrying junk deep into the funnel.
- Leaking at qualified-to-demo? Deals are qualifying and then stalling before serious evaluation. That usually points at discovery that didn't establish enough urgency, or follow-up that let warm deals go cold in the gap.
- Leaking at proposal-to-negotiation? Your offer or demo isn't landing, or — very often — price is the unspoken wall, which is its own negotiation problem rather than a selling one.
- Leaking at negotiation-to-close? Deals that should close are dying at the finish line — usually a closing discipline gap or an unhandled objection. A win/loss analysis on these tells you which.
The same overall win rate can come from wildly different leak patterns, and each one calls for a completely different fix. The number that matters isn't the average — it's the shape.
Conversion rate and velocity are two halves of the same picture
A leak isn't only about whether deals advance; it's also about how long they take to. A stage where deals technically convert but sit for six weeks is leaking in slow motion — those deals tie up capacity, go stale, and quietly die of stall even though they never officially "lost." That's why conversion rate is the natural partner of sales velocity: one tells you where deals fall out, the other tells you where they get stuck. Read together, they point at the same problem stages from two angles, and a fix that improves one usually improves the other.
This is also why a low conversion rate at a gate isn't automatically the place to intervene. If a stage converts at thirty percent but only ten deals pass through it a month, fixing it returns little. If a stage converts at sixty percent but five hundred deals flow through, a few points of improvement there dwarfs everything else. Weight every leak by the volume behind it, and chase the gate where conversion rate and throughput together represent the most recoverable revenue.
See your funnel without doing the math by hand
You cannot manage stage conversion you can't see, and computing it from a spreadsheet export each month is exactly the chore that gets skipped until a bad quarter forces a fire drill. The fix is to make the funnel a live picture, not a periodic project.
In Hitt CRM, reports compute conversion rate at each stage straight from your real pipeline, so the leakiest gate is visible at a glance instead of reverse-engineered after the fact. Because the funnel reads from the same records reps work every day, the numbers stay honest as long as the pipeline stays clean — and you can watch a conversion rate move after you change something, which is the only way to know whether a fix actually worked. The funnel stops being a quarterly autopsy and becomes a dashboard you steer by.
The one-sentence version
A single win rate hides the only fact worth knowing — which stage your deals actually leak out of — so measure conversion rate gate by gate, read the pattern (each leak points at a different broken thing, from targeting to discovery to closing), weight every leak by the volume behind it, pair it with velocity to catch deals that stall rather than lose, and watch it live so you fix the gate that's truly costing you instead of the one you assumed.