How should reporting work across multiple locations or brands?
Build it bottom-up: identical metric definitions at every location, then a rollup that aggregates them correctly. The rollup is for comparison and resource allocation; the location view is for management. The hard part is not the math — it is getting every location to tag jobs, sources and service types the same way, because inconsistent tagging makes location comparison meaningless.
Tagging discipline is the actual blocker
Every multi-location reporting project runs into the same wall: location A calls it a maintenance visit, location B calls it a tune-up, location C files it under service. The moment you compare service-line mix across them, the differences you see are naming conventions rather than business differences.
Fix this before you build anything. Either standardize the tags at the source or maintain an explicit mapping table in the integration layer. A mapping table is usually faster to deploy and is honest work — it is a normal part of systems integration, not a shortcut.
Rollups are not averages of averages
A company-wide close rate is total closes divided by total opportunities, not the mean of location close rates. Those two numbers differ, sometimes substantially, and the gap grows as locations differ in size. The same trap applies to average ticket, cost per booked job and anything else expressed as a ratio.
The rule: roll up numerators and denominators separately, then divide. If your reporting shows a rollup ratio, ask which way it was computed.
Normalize before you rank
- Per tech day. Revenue per tech day makes a six-truck shop comparable to a twenty-truck shop.
- Per capita or per household in the service area. Market size explains much of what looks like performance difference.
- Per dollar of media — expressed as cost per booked job — rather than raw lead volume.
- Same service mix. A location weighted toward replacements will always look better on average ticket and worse on job count.
Report spread, not just the total
The most useful multi-location view is not the aggregate. It is the distribution: which locations sit outside the normal band, on which metric, and by how much. The total hides exactly the information the rollup exists to surface.
Then make outliers actionable in both directions — the bottom location needs help, and the top one usually knows something worth copying. Comparative analysis across locations is one of the few places where automated pattern-finding beats a person reading a table.
The first months mostly surface data quality
Set expectations before launch: the first rollups will mainly reveal that locations record things differently. Somebody dispatches differently, somebody else leaves half the source fields empty, a third has been using a custom job type since before the acquisition.
This looks like the project failing and is actually the project working — those inconsistencies were always there, they were just invisible. Budget a cleanup phase, fix the worst offenders at the source, and treat the first comparable quarter as the real starting line rather than the launch date.
Topics: multi-location · rollup reporting · comparison · franchise
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.