The problem with sales KPIs isn't finding them — every CRM will happily chart forty. The problem is that only a handful change what you do on Monday, and the rest are decoration that makes dashboards feel productive. Here are the fifteen worth tracking, organized by the question they answer, plus five that routinely mislead.
Top of funnel: is enough coming in?
1. Qualified leads per period — not raw leads; leads passing your MQL/SQL bar. Raw lead count is how bad channels hide.
2. Lead response time — minutes from lead arrival to first meaningful touch. The most controllable, highest-leverage number on this list.
3. Cost per qualified lead, by source — spend divided by qualified output per channel. Kills expense-center channels that flood you with junk.
4. Lead-to-opportunity conversion — what fraction of qualified leads become real deals; the honesty check on your definition of "qualified."
Middle of funnel: is the machine converting?
5. Stage-to-stage conversion rates — where the funnel actually leaks, so you fix the right stage (details in our CRM reports guide).
6. Pipeline coverage ratio — open pipeline ÷ target; compute your required multiple from your own win rate, not the 3x folklore (full math here).
7. Average deal age by stage — the zombie detector. Deals past normal dwell time are usually dead-but-unmarked.
8. Follow-up persistence — average touches per open opportunity. Almost nobody tracks it; it predicts win rate better than most things teams do track.
Bottom of funnel: are we winning?
9. Win rate — pick one definition (wins ÷ decided, or wins ÷ all) and never switch mid-year.
10. Average deal value — watched jointly with win rate; discounting your way to wins shows up here.
11. Sales cycle length — median days from open to close; most "long cycles" are mostly dead air between touches.
12. Sales velocity — the composite of 9, 10, 11 and opportunity count; the single best trend metric in sales (formula and levers).
After the close: does revenue stay?
13. Forecast accuracy — forecast vs. actual, monthly. A consistent one-direction gap is a calibration error, not luck.
14. Win/loss reasons coded — a KPI about discipline: what fraction of closed deals carry an honest reason code. Below ~90%, your loss analysis is fiction.
15. Net revenue retention (where applicable) — expansions minus churn on the existing base; the metric that decides whether new sales are filling a bathtub with the drain open.
Five metrics that lie
- Total pipeline value. Inflated by zombies and wishful sizing; only meaningful next to aging and coverage math.
- Activity counts as targets. Useful for diagnosis (effort vs. effectiveness), corrosive as quotas — you'll get impressive dials and empty calls.
- Email open rates. Inflated and privacy-blurred; replies are the real signal (our subject-line guide says the same).
- Raw lead volume. The favorite metric of channels that don't convert.
- Quota attainment alone. One rep at 110% on huge discounts and another at 95% on clean deals — attainment ranks them backwards.
Making KPIs operational
- Assign each KPI a rhythm and an owner — daily (response time, at-risk deals), weekly (pipeline, aging, persistence), monthly (conversions, forecast accuracy), quarterly (velocity, win/loss, NRR).
- Track trends, not absolutes. Your numbers against your last quarter beat anyone else's benchmarks — context differs too much for borrowed targets.
- Fix the data layer first. KPIs computed on hand-logged data measure logging discipline, not sales. Automatic activity capture is the prerequisite — in AutomateNexus CRM the agents log touches, age deals, and recompute the forecast continuously, so the numbers describe reality without anyone maintaining them.
Sales KPI FAQ
What are the most important KPIs for a small sales team?
Start with four: lead response time, follow-up persistence, win rate, and sales velocity. They cover speed, discipline, quality, and the composite — everything else can wait until those are healthy.
How many KPIs should a sales team track?
Actively manage 5–8; review the rest on a cadence. Past that, dashboards stop changing behavior and start decorating meetings.
What's the difference between a KPI and a metric?
A metric is anything you can measure; a KPI is a metric tied to a decision and an owner. The test: if this number moved 20%, would anyone do something different? No = metric, not KPI.