A multi-location brand runs the same marketing meeting twice. Corporate looks at the rollup and asks why one region is slipping. A store manager looks at her own numbers and says she just had her best month of the year. Both are sure. Neither trusts the other's report.
That gap is not a personality problem. It is a reporting problem. We have built reporting for multi-location brands, and the failure pattern is the same almost every time. So is the fix.
Why nobody trusts the report
Multi-location reporting usually grows up in pieces. Corporate builds a dashboard from the ad platforms. Each location tracks what it can see, usually foot traffic, phone calls, and register sales. The two sets of numbers were never designed to match, so they never do.
Then definitions drift. "Leads" means form fills in one report and phone calls in another. "This month" means calendar month in one export and trailing thirty days in the next. Small differences, big arguments.
Once a manager catches one mismatch, every future report is suspect. Trust does not degrade gradually. It breaks once and stays broken.
Comparability comes before accuracy
The instinct is to chase perfect attribution first. Resist it. The first job is making locations comparable to each other, because comparison is what corporate actually does with the report.
Comparable means three things.
- Same definitions. Every metric on the report has one written definition that applies to every location. No local variants.
- Same date logic. Every location's numbers cover the same window, pulled the same way, on the same schedule.
- Same source. One dataset feeds every view. Nobody maintains a private spreadsheet on the side.
Then compare on what a manager can control. Cost per lead is heavily shaped by the local market. Grading a downtown metro store against a rural one on that metric punishes geography, not effort. Judge each location against its own baseline first, and save cross-location rankings for metrics where the playing field is level.
Naming is the load-bearing wall
Here is the unglamorous truth: every automated rollup depends on campaign names.
The reporting system has to know that a given campaign belongs to store 41, sits in the Southeast region, runs on search, and promotes store traffic. If that information is not encoded in the name, a human has to map it by hand. Hand mapping means lag, errors, and a report that quietly rots.
A workable convention encodes region, location, channel, and objective in a fixed order. Something like:
SE_TX-NB-041_Search_Brand_StoreTraffic
The exact format matters less than two rules. Every campaign follows it, and violations get caught fast. An automated check that flags a nonconforming name the day it launches is worth more than any quarterly cleanup. This is part of what we wire in when we build paid ads systems. Naming is not housekeeping. It is the foundation the reporting stands on.
Rollups for corporate, drill-downs for locations
Corporate and location managers do not need the same report. They need the same data shaped two ways.
Corporate needs the rollup. Totals by region, trends over time, budget pacing, and a short list of outliers worth a conversation. Ten minutes of reading, then decisions.
A location manager needs the drill-down. Her store, her campaigns, her trade area, in plain language. What ran, what it cost, what it produced, and how that compares to her own recent history.
The critical property: the drill-downs must sum to the rollup. Exactly. When they do, the two-meetings problem disappears, because anyone can trace their number into the other view. When they do not, you are back to dueling spreadsheets.
The anti-pattern is two teams building two reports from two exports. One pipeline, one dataset, two views. That is the whole architecture.
Summaries that write themselves
A dashboard is not communication. Most location managers will not log into one, and asking corporate to write a weekly narrative for hundreds of locations is a nonstarter. So most locations get silence.
This is where AI earns its place in the engine. Once the data layer is clean and the naming holds, a short written summary for each location can generate itself every week. What moved, the likely reason, and what changes next week. Three sentences a manager actually reads, with the source numbers cited so she can check the math.
The summary is only as trustworthy as the layer beneath it. Automate a narrative on top of messy data and you scale the mistrust. That order of operations, data first and narrative second, is the core of how we build analytics and reporting.
When location managers push back
Pushback is not a problem to manage. It is free quality assurance.
When a manager says the report is wrong, one of three things is true. The data is wrong, the definition was never communicated, or the report is missing context it should carry. All three are fixable. All three are worth finding.
A few rules make pushback productive instead of political.
- Publish the definitions. Every metric, in writing, on one page, linked from the report itself.
- Show the lineage. A manager should be able to see which campaigns and which date range produced her number.
- Give challenges a lane. A named place to flag a suspect number, and a committed turnaround for the answer.
- Hold rankings back at first. Run the first stretch as review only, so managers can argue with the numbers before the numbers carry consequences.
- Fix the first real error loudly. When a manager catches a real mistake, correct it, announce it, and credit her. Nothing builds trust in a system faster than watching it take a correction well.
The layer underneath the argument
Corporate versus locations is rarely a data fight. It is a fight about whose version of reality counts, held on top of a reporting layer that was never built to produce one version. Build that layer and the fight has nothing to stand on.
This is marketing systems work. If your rollups are hand-built, your naming is inconsistent, or your managers keep private spreadsheets because they do not trust yours, look at the whole operational layer at once instead of patching one dashboard. That is what our marketing operations audit covers, and it is where the Performance Audit begins: a fixed-scope review that maps where the reporting breaks, with a dollar-weighted plan you keep either way.