How do I read Google Business Profile performance data without fooling myself?
Split every metric by whether the searcher already knew you. Searches that used your business name are demand you created elsewhere; searches that used a category term are discovery. Total views mixing both will rise whenever your brand advertising works and tell you nothing about local search. Track discovery searches and profile-originated calls separately, and compare them to booked jobs rather than to each other.
The number everyone quotes is the least useful one
Profile views is a compound metric. It rises when your radio spot runs, when a truck wrap gets noticed, when a competitor goes out of business, and occasionally when your local SEO improves. Reporting it as a local SEO result is not exactly dishonest, but it is not evidence either.
The first cut that makes the data usable is branded versus unbranded. A search for your company name is a person who already decided to look you up. A search for a category term is a person choosing between options, and only the second group is the audience local search work is trying to reach.
Which profile metrics carry real information
- Discovery or category searches. The closest available proxy for whether you are being found by people who did not know you.
- Calls from the profile. An action, not an impression. This is the metric worth defending in a meeting.
- Direction requests. Meaningful for storefronts, close to noise for service area businesses.
- Website clicks. Useful mainly as a denominator against what happens after the click.
- Booking or message interactions, where enabled, which usually behave like calls and should be counted alongside them.
Three traps in the reporting
Seasonality dominates. Home services demand swings hard enough that month-over-month comparisons are mostly weather. Compare to the same month last year, or to a rolling twelve-week average, and say which you used.
Attribution windows differ. Profile data is reported against the interaction date. Your revenue is recorded against the job date. A call in one month that becomes a job in the next will not line up unless someone deliberately joins them.
Sampling and definition changes happen. Google adjusts what it counts and how far back data goes. A step change on a specific date is usually a definition change, not a performance change, so annotate the date rather than build a theory around it.
The join that turns it into a business metric
On its own, profile data tops out at telling you that activity happened. To know what it was worth, the calls originating from the profile have to be identifiable and traceable into the operational system, so a profile call can be followed to a booking and to an invoice.
That is the ordinary work of connecting marketing surfaces to the field service record, and once it exists the question changes from how many views did we get to what did local search produce in booked revenue. See marketing intelligence for how that join is structured and revenue intelligence for what it lets you report. Without it, local search stays the one channel everybody defends with impressions.
Topics: performance data · branded search · reporting · measurement
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.