How do you compare locations fairly when they are different sizes and markets?
Normalize by capacity and demand instead of ranking raw totals. Revenue per technician-day, booked jobs per available slot and cost per booked job travel across markets far better than revenue or lead volume. Then rank managers on the metrics they actually control, like booking rate, on-time arrival and estimate close rate, and keep market-driven figures on a separate panel.
Separate what a manager controls from what the market hands them
A location in a dense, affluent, high-demand market will out-earn a rural one forever. Ranking on revenue teaches nothing and demotivates the people doing the best work with what they have. Split the scorecard: one panel of controllables, one panel of conditions.
Controllables are things a manager can change this month. Conditions are things that take a year or a budget to change. Mixing them into one leaderboard guarantees the ranking mostly measures geography.
A useful test for whether something belongs in the controllable column: could a competent manager move this number within a quarter using resources they already have? Lead volume usually fails that test. Booking rate on the leads they receive usually passes it.
Normalizers that hold up across markets
- Per technician-day. Revenue and completed jobs divided by actual working technician-days removes headcount differences and vacation effects.
- Per available capacity hour. Booked jobs against schedulable capacity separates a demand problem from a staffing problem.
- Per household in the service radius. A crude but useful denominator for comparing market penetration rather than market size.
- Weather-normalized. In heating and cooling, degree-days explain a large share of demand variance between regions; comparing raw call volume across climates compares climates.
Rank on trend, not on level
A structurally disadvantaged location can be the fastest improving one in the company. Level rankings hide that and level rankings never change, so people stop reading them. Rank on change against the location's own trailing baseline and the list becomes informative every month.
Keep the level visible as context, but let the sort order reward movement. This is also the ranking that identifies practices worth copying, because it points at what recently started working somewhere.
Trend ranking has a second benefit. It surfaces decline early at strong locations, which level rankings hide completely because a top performer sliding toward average still sits near the top of the list for months.
A rollup is only as good as its least consistent location
Cross-location comparison fails most often on data entry, not on analysis. One location codes membership work as maintenance, another books estimates as jobs, a third uses a different lead source list. The report then compares vocabulary, not performance.
Before publishing any comparison, verify that job types, statuses, lead sources and date bases mean the same thing everywhere. That normalization is the real work in field service system integration, and it is the prerequisite for any credible multi-location intelligence.
Run a consistency check before every comparison cycle: count distinct job types, lead sources and statuses actually used per location. Divergent counts are the fastest signal that two locations are recording the same work differently.
Topics: multi-location · benchmarking · normalization · capacity
Have a version of this question about your own business?
The useful answer usually depends on which systems you run and how they're connected. That's a conversation, not a blog post.