SDR KPIs worth tracking, and the denominators that ruin them
The handful of SDR metrics that actually drive decisions, how each one gets gamed, and the measurement mistakes that make a dashboard confidently wrong.
Held meetings
58
vs 52 same point last month
Show rate
79%
58 held of 73 booked
Working days
19
weekdays minus holidays
There is no shortage of SDR metrics you can track. The useful question is which ones change a decision. This is a framework, not a benchmark table — we are not publishing industry averages, because the ones circulating are mostly unsourced and a made-up number is worse than none.
Four SDR KPIs that drive decisions
Held meetings per rep
The output metric. Held, not booked.
Show rate
Held divided by total booked. The honesty check on the first metric, and usually the most actionable number.
Bookings per working hour
Per rep, per client. The only way to answer "can we take this client on in six weeks" with arithmetic.
Money per hour
Revenue divided by hours, per rep and per client. Decides whether a client is worth keeping — and is rarely computed.
Everything else — dials, connects, conversation rate, email reply rate — is diagnostic. Useful for working out why the four moved, not for deciding anything on its own.
How each metric gets gamed
Any metric that drives pay will be optimised, including in ways you did not intend:
| Metric | The predictable distortion |
|---|---|
| Bookings | Low-quality meetings booked to hit a number |
| Dials | Short, pointless calls to inflate the count |
| Connects | Calling known-easy contacts repeatedly |
| Show rate | Reluctance to book anything marginal — real pipeline left on the table |
| Money per hour | Under-logging hours |
Two guards help. Pay on held rather than booked, which removes the main volume incentive. And take hours and outcomes from the system rather than self-report — logged activity and recorded outcomes are much harder to shade than a timesheet.
Wrong denominators break most dashboards
Most broken dashboards use the right metrics with the wrong denominators.
Wrong denominator
- Weekdays — December looks like a collapse next to March
- Booking month and meeting month mixed on one chart
- A reschedule counted as a new booking
- One team average across clients with different list quality
- Two bookings in three hours treated as a rate
Honest denominator
- Working days: weekdays minus public holidays, absences and part-time schedules
- One month rule, applied everywhere
- A reschedule is the same booking with a new date
- Rates per rep and client, aggregated only when the mix is known
- Start from a prior and let the estimate move with evidence
Working days. Counting weekdays makes December look like March, so per-day productivity appears to collapse in holiday-heavy months and the team gets a performance conversation it did not earn. Use the public holidays of the country the account operates in.
Booking month vs meeting month. A meeting booked on 29 March for 4 April: booked-month flatters March and starves April; meeting-month is usually the honest choice for show rate but makes the booking rep’s output appear late. Pick one and never mix them.
Small samples. A rep with three hours on a new account does not have a bookings-per-hour rate. Planning against 0.67/hour from two bookings is how forecasts embarrass people.
Leading and lagging indicators
Activity metrics are leading indicators, and weak ones. A rep can make the dials and produce nothing. Their real value is diagnostic: when held meetings drop, activity tells you whether the cause is less work or worse conversion.
The more useful leading indicator is list health — how much workable list remains, and what share of its numbers connect. A list running dry predicts next week’s output better than this week’s dial count.
What to put on the team lead’s dashboard
Four rows and one chart, as in an SDR dashboard:
- Held meetings this period vs the same point last period
- Show rate, with the booked count beside it
- Remaining workable list per client, with a depletion warning
- Money per hour per client, ranked
Then one chart: held meetings by week, per client, so you see the shape rather than a single number. Everything else belongs one click down, for diagnosing rather than deciding.
Benchmark against your own numbers
Teams often ask what a good show rate or bookings-per-hour figure is. Your own trailing numbers, segmented by client, are a better benchmark than any industry figure — they already control for your market, list quality, pricing and qualification bar. Measure yourself against last quarter, not against a blog post.