Measuring operational leverage, not output
Output metrics reward busywork. Leverage metrics reward systems that compound. Here are the four we instrument first on every engagement.
Walk into most operations dashboards and you'll see a wall of output metrics: tickets closed, calls handled, orders processed, units shipped. They're easy to count, easy to chart, and almost completely useless for telling you whether the operation is getting better.
Output metrics reward activity. Leverage metrics reward systems. The difference shows up over quarters, not weeks — which is why most teams never make the switch.
1. Cycle time
From the moment a customer expresses intent to the moment that intent is resolved. Shrinking this number is almost always net positive — customers are happier, the team handles more volume per person, and downstream effects (refund disputes, churn, repeat contacts) all improve.
Measure end-to-end, not per-step. A team that's fast at step 2 and slow at step 5 still has a slow cycle time. Per-step metrics hide where the actual problem is.
2. Repeat-contact rate
How often the same customer comes back about the same issue within 7 days. This is the single most underused metric in operations. It's a direct proxy for resolution quality — a closed ticket that generates a follow-up wasn't really closed.
Teams that ignore repeat-contact rate optimize for ticket closure speed and accidentally make the underlying problem worse. A 'fast' team with a 40% repeat rate is doing the same work three times. A 'slow' team with a 5% repeat rate is actually faster.
3. Automation share
What percentage of work flows through the system end-to-end without a human touching it. Not 'tickets deflected by the bot' — full end-to-end resolution including any backend actions. Track the trend, not the absolute number.
A rising automation share means the system is absorbing complexity that used to require human judgment. A flat or declining share means the operational debt is growing faster than your investment in automation. Either way, the trend tells you whether you're winning the long game.
4. Operator hours per unit
Total team hours divided by units of work completed (tickets, orders, onboardings — whatever the operation produces). This is the leverage metric. When it goes down, you've genuinely gotten more efficient. When it stays flat while volume grows, you're scaling linearly — and you'll hit a wall.
"If your headcount has to grow at the same rate as your volume, you don't have an operation. You have a queue."
Why this set, and not others
These four metrics resist gaming. You can't fake a lower cycle time by closing tickets faster — repeat-contact rate catches it. You can't fake automation share with theatre — operator hours per unit catches it. Together they describe whether the operation is compounding or just running. That's the only question that actually matters at scale.