The Two-Number Split: technician efficiency vs productivity, and what each one tells you

The two-number split is two measures read side by side, in the same order as the video:

1. Efficiency. Hours billed against the hours the tech spent on the job. It measures speed at the work.
2. Productivity. Hours billed against the hours the tech was clocked in. It measures how much of the day became billable work.

Read them together. When efficiency looks fine and productivity is low, the tech isn't slow. The hours went to waiting. Below are the formulas, a worked example with demo numbers, the list of where those hours usually go, and how to talk about the numbers without turning them into a stopwatch.

Who this is for

Owners and service managers who feel their techs should be producing more and want to know whether the fix is coaching or clearing the way. You need two or three weeks of clock-in data and billed hours per tech from your shop system.

Plenty of shops don't look at either number. In its 2024 industry survey, Ratchet+Wrench found that 18% of shops did not track technician efficiency and 22% did not track productivity. If that's you, this is the place to start.

Why it's worth the hour

Techs are hard to replace. The Bureau of Labor Statistics counts about 825,800 automotive service technicians and mechanics in 2025 and about 66,200 openings a year, many of them from people leaving the job or retiring. If hiring another tech is slow, more finished work from the techs you have is the other lever, and that lever is mostly about their time between jobs.

The two formulas

Many shops use these working definitions. Check how your own shop system defines them before you compare with anyone else, because vendors differ.

Measure	Formula	What it tells you
Efficiency	Hours billed on jobs ÷ hours the tech actually spent on those jobs	Speed at the work
Productivity	Hours billed ÷ hours clocked in	Share of the day that turned into billable work

A worked example

These numbers are made up for a demo tech, Luis, over one week.

• Clocked in: 40.0 hours
• Time actually on jobs: 31.0 hours
• Hours billed: 31.5 hours

Efficiency is 31.5 divided by 31.0, a little over 100 percent. Luis is fine at the work. Productivity is 31.5 divided by 40.0, which is about 79 percent. About nine hours of his week didn't turn into billable work, and his speed had nothing to do with it.

How to read the split

• Efficiency low, productivity fine. The tech is on jobs most of the day but takes longer than the billed time. Look at the job: skill, tools, a hard diagnosis, or a flat-rate time that doesn't fit the repair. This is a coaching or training question.
• Efficiency fine, productivity low. The tech works at speed when there's work in front of them, and there often isn't. This is a workflow question. Look at the waits.
• Both low. Start with the waits anyway. They're usually faster to fix, and fixing them often lifts both.
• Both high. Leave it alone and ask the tech what helps.

Where the hours usually go

Shop4D, a shop-software company, lists wasted tech time as a common bottleneck: techs waiting on answers or parts and losing billable time. AutoVitals makes a similar point, saying much technician downtime traces back to earlier workflow gaps, not scheduling. Both are vendors without a measured figure, but the list matches what most owners see:

1. Inspection done, no estimate yet. The next job isn't approved, so the tech waits or picks up busywork.
2. Approved job, part not at the bay. The tech walks to the counter, or waits.
3. A question for the advisor, and the advisor is on the phone.
4. Notes and photos entered after the fact, because the tech was in the middle of something else.
5. Rework from a finding missed on the first look.
6. No clear next job. The board doesn't say what's next, so the tech asks.

For a week, have the tech or the dispatcher jot down each wait over ten minutes with its cause. Add them up. The biggest cause is your first fix.

What to fix first

• Estimates waiting. Set a target time from inspection done to estimate sent, and standardize how findings are written.
• Parts not at the bay. One person owns parts status, and every ordered part gets an ETA. A job with no ETA gets chased before the tech starts it.
• Questions waiting. Give techs one way to flag a question that the advisor sees without a phone call.
• No next job. Keep a visible queue of approved work with parts in hand.

Software can help with pattern finding. Shopmonkey's 2024 post lists AI analytics that identify repairs taking longer than expected. Used well, that shows which repair types or bays run long. Used badly, it ranks people.

How to talk about the numbers

• Tell the team what you measure and why before you start. Say plainly it's about the waits.
• Share each tech's numbers with that tech.
• Watch for flat-rate distortions. A job billed at two hours and finished in just over one reads as far above 100 percent efficient, which can hide a rushed repair or a padded book time. Treat any extreme number as a reason to talk, not a score.
• Keep coaching, pay and dispatch decisions with people.

How to tell it worked

Run the split for two weeks before any change and two weeks after. If you fixed the waits, productivity should rise while efficiency stays about where it was. If productivity didn't move, the fix hit the wrong wait, so go back to the list and pick the next one.

The longer explanation is on technician productivity AI for auto repair. For a two-week way to tag the waits by cause, see auto shop bottleneck detection. RevvOS's daily operations view shows delayed parts and aging repair orders so someone can clear them before a tech is waiting.