Ask a local business where its website visitors come from and the answer is usually "Google". Ask how many of those came from its Business Profile rather than the ordinary organic results, and the room goes quiet.
The two are easy to blur. In GA4, untagged visits from a Business Profile's website button often look just like visits from a normal search result: google as the source, Organic Search as the channel. If you want to know whether the profile is pulling its weight, you need to separate them, and you need the profile's own interaction data alongside your analytics.
This post covers both halves: tagging profile links so GA4 can isolate them, and pulling calls, website clicks and direction requests from the Business Profile Performance API.
Part 1: Tag the links on your Business Profile
A Business Profile can carry a website link and, for some business types, an appointment or booking link. Add UTM parameters to each so the sessions they create are identifiable.
https://www.example.co.uk/?utm_source=google&utm_medium=organic&utm_campaign=gbp-listing&utm_content=website-button
https://www.example.co.uk/book/?utm_source=google&utm_medium=organic&utm_campaign=gbp-listing&utm_content=appointment-link
Why utm_medium=organic and not something like gbp or referral? Because of how GA4 builds its default channel groups. Google's default channel group definitions say Organic Search matches when the source is on its list of search sites or when the medium exactly matches organic. Referral matches when the medium is referral, app or link.
So utm_medium=organic keeps profile traffic where it belongs conceptually, inside Organic Search, while utm_campaign=gbp-listing lets you pull it out whenever you need to. A custom medium such as gbp would push those sessions into an unhelpful bucket, and referral would make your Business Profile look like just another website linking to you.
Two habits keep this clean:
-
Lowercase everything. Google's guidance on collecting campaign data with custom URLs recommends a strict, case-sensitive naming convention and lowercase as standard.
utm_source=googleandutm_source=Googleare different values. -
Use one campaign name across every location. If you manage several branches, keep
utm_campaign=gbp-listingconstant and useutm_contentor a separate landing page to tell branches apart. Consistency is what makes the report usable later.
Check the tagged URL before saving it
A redirect that strips query strings will quietly break all of this. Test the exact URL before you paste it into the profile:
curl -sIL "https://www.example.co.uk/?utm_source=google&utm_medium=organic&utm_campaign=gbp-listing&utm_content=website-button" | grep -i "^location"
If any Location header drops the parameters, fix the redirect, typically a forced trailing slash, an HTTP-to-HTTPS rule or a www rewrite, so query strings are preserved.
Part 2: Find profile traffic in GA4
Once the tagged link has been live for a while, the sessions are easy to isolate:
- Open Reports → Acquisition → Traffic acquisition.
- Change the primary dimension to Session campaign.
- Look for
gbp-listing. Add Session manual ad content as a secondary dimension to split the website button from the appointment link.
For anything more than a glance, build an exploration: rows for session campaign and landing page, columns for device category, and your key events (enquiry form submissions, generate_lead, bookings) as values. That shows whether profile visitors actually do anything once they arrive.
A useful comparison is profile sessions against all other Organic Search sessions for the same key events. If profile visitors convert at a very different rate, that is a prompt to look at the landing page you are sending them to. Many businesses send profile clicks to a generic homepage when a location or service page would answer the visitor's question faster.
Part 3: Pull interaction data from the Performance API
GA4 only sees people who reach your site. A Business Profile also generates calls, direction requests and other actions that never touch the website. The Business Profile Performance API exposes those as daily time series.
The endpoint is a GET request:
GET https://businessprofileperformance.googleapis.com/v1/locations/{locationId}:fetchMultiDailyMetricsTimeSeries
It needs the OAuth scope https://www.googleapis.com/auth/business.manage, repeated dailyMetrics parameters and a dailyRange with start and end dates. The documented daily metrics include CALL_CLICKS, WEBSITE_CLICKS, BUSINESS_DIRECTION_REQUESTS, BUSINESS_BOOKINGS and impressions split by Maps or Search and by desktop or mobile.
Here is a small Node script (Node 18 or later, saved as an .mjs file) that totals three metrics for a date range:
// gbp-metrics.mjs
const LOCATION_ID = process.env.GBP_LOCATION_ID; // numeric listing ID
const TOKEN = process.env.GBP_ACCESS_TOKEN; // OAuth token with business.manage scope
const METRICS = ['WEBSITE_CLICKS', 'CALL_CLICKS', 'BUSINESS_DIRECTION_REQUESTS'];
function buildParams(start, end) {
const params = new URLSearchParams();
for (const metric of METRICS) params.append('dailyMetrics', metric);
params.set('dailyRange.start_date.year', start.year);
params.set('dailyRange.start_date.month', start.month);
params.set('dailyRange.start_date.day', start.day);
params.set('dailyRange.end_date.year', end.year);
params.set('dailyRange.end_date.month', end.month);
params.set('dailyRange.end_date.day', end.day);
return params;
}
const params = buildParams({ year: 2026, month: 10, day: 1 }, { year: 2026, month: 10, day: 31 });
const url = `https://businessprofileperformance.googleapis.com/v1/locations/${LOCATION_ID}:fetchMultiDailyMetricsTimeSeries?${params}`;
const res = await fetch(url, { headers: { Authorization: `Bearer ${TOKEN}` } });
if (!res.ok) throw new Error(`API error ${res.status}: ${await res.text()}`);
const data = await res.json();
for (const multi of data.multiDailyMetricTimeSeries ?? []) {
for (const series of multi.dailyMetricTimeSeries ?? []) {
const values = series.timeSeries?.datedValues ?? [];
// value is returned as a string and omitted on zero days
const total = values.reduce((sum, point) => sum + Number(point.value ?? 0), 0);
console.log(`${series.dailyMetric}: ${total}`);
}
}
Two details in that loop matter. The reference for TimeSeries documents value as a string, so convert it before adding. It also notes that value is omitted when it is zero, so default missing values to zero rather than treating them as errors.
For authentication, use whichever OAuth flow fits your setup. A server-side job usually stores a refresh token for an account that manages the profiles and exchanges it for access tokens on each run. Follow Google's Business Profile API setup documentation for project access before you expect the first call to succeed.
Part 4: Put the two views side by side
Once you have both data sources, resist the urge to make them match. They measure different things.
| Signal | Source | What it tells you | What it cannot tell you |
|---|---|---|---|
WEBSITE_CLICKS |
Performance API | Clicks on the profile's website button | Whether the page loaded or the visitor stayed |
gbp-listing sessions |
GA4 | Visits that reached your site and were measured | Visitors who declined analytics consent or blocked tags |
CALL_CLICKS |
Performance API | Taps on the call button | Whether a call connected or was answered |
BUSINESS_DIRECTION_REQUESTS |
Performance API | Direction requests | Whether anyone arrived |
Key events from gbp-listing
|
GA4 | Enquiries and bookings from profile visitors | Offline outcomes such as completed jobs |
Expect GA4 sessions to be lower than API website clicks. Consent choices, blocked scripts, bounced page loads and different counting rules all widen the gap. Track the ratio over time instead: a sudden drop in sessions relative to clicks often means a broken redirect or a tagging mistake rather than a fall in interest.
A monthly summary that combines both is usually enough for a small business: profile website clicks, profile sessions, enquiries from those sessions, call clicks and direction requests. For multi-location brands, the same table per location quickly shows which branch profiles need attention.
Common mistakes
-
Tagging with mixed case (
utm_source=Google), which splits the data. -
Using
utm_medium=referral, which moves profile traffic out of Organic Search. - Sending profile clicks to the homepage when a location or service page would serve the visitor better.
- Treating a call click as a call. It is an intent signal; your phone system holds the real outcome.
-
Forgetting the appointment link. If the profile has one, tag it with its own
utm_contentvalue.
If you want the wider context for these numbers, the measurement section of our local search guide explains why customer actions matter more than a single ranking position.
Prox Digital Agency builds and measures websites for local and service businesses; this setup is the baseline we use before discussing any local search results.
This post was drafted with AI assistance.
Top comments (0)