Skip to content

Five Google Business Profile Findings from 45,000 Listings

This guide distils a 2019 study of 45,000 Google My Business listings (now Google Business Profile) into five findings a local marketer can act on. The figures are a historical reference point, not a current benchmark — check present-day conditions before treating any number as today's norm.

What the study covered

The analysis looked at 45,000 listings across 36 industries over a year and a half, producing more than 8 million data points. It examined how consumers use the platform, how behaviour shifted over time, which industries led and lagged, and what benchmarks could help a business judge its own performance. The five findings below are the headline takeaways.

1. A website still earns its place

Across nearly every industry and country, website clicks were the most common customer action recorded in GMB Insights. A separate study referenced in the same analysis found that 76% of consumers say they look at a website at least half of the time when researching a local business. The practical reading: a well-linked website remains a primary conversion path from the listing, so keep the link accurate and the landing page relevant to the search that surfaced the listing.

2. Photo count tracks with engagement

The average listing carried 11 photos. Comparing photo counts with customer actions showed a clear pattern: more images went with more actions. Listings with over 100 images drew 20 times more website clicks than the average listing, while listings with a single image generated 65% fewer clicks than average. About 6% of businesses had only one image.

The study was careful to note this is correlation, not proven causation — businesses with many photos may simply be stronger at local marketing overall. Still, adding high-quality images is a low-risk optimisation, and the checklist below turns the finding into a routine.

3. Most views come from discovery, not brand searches

84% of listing views came from discovery searches — non-branded queries such as "bars near me", "dentist in brooklyn" or "greek restaurant". That makes category selection and keyword-relevant listing setup central to visibility, not an afterthought. If your listing only performs when someone already knows your name, the discovery share is the gap to close.

4. The customer action rate rose year on year

The share of listing views that produced a customer action moved from 3.9% in 2017 to 4.8% in 2018. The study's working explanation was that new platform features gave searchers more reasons and opportunities to engage, and that consumers grew more comfortable using them. Treat the two percentages as a historical comparison; track your own action rate against your own baseline rather than assuming the 2018 figure applies now.

5. Car dealerships stood out

Industry-level optimisation and engagement varied widely. Frequently visited businesses such as bars and restaurants naturally attract more searches, and holiday-driven photo habits help explain why the average hotel had over 13 times more images than the average business. The surprise was car dealerships: third for average photo count, second for direction requests, and top for calls and website clicks. Buying or leasing a car is a high-ticket, low-frequency purchase, yet dealership listings tended to be well optimised and the sector appeared to have embraced the platform early.

A checklist drawn from the five findings

  • Confirm the website link on your listing resolves to a relevant, fast landing page.
  • Count your photos. If you are near the single-image end, add a batch of genuine, high-quality images and keep adding over time.
  • Review your primary and secondary categories against the discovery-style queries you want to appear for.
  • Record your own views-to-action rate monthly so you can see your trend rather than relying on the 2017–2018 figures.
  • Compare your photo count, calls, direction requests and website clicks with a same-industry peer before drawing conclusions.
  • Remember the photo finding is correlational: pair image work with broader listing hygiene rather than treating photo count alone as the lever.

Worked example: reading the photo numbers

Suppose your listing has 11 photos — the study average. The click contrast in the data is between the average listing and two extremes: over 100 images (20× the average website clicks) and a single image (65% fewer clicks than average). The useful move is not to chase 100 images blindly, but to treat photo count as one signal among several. Add images in planned batches, then watch whether website clicks and calls move alongside them. If they do not, the constraint is likely elsewhere — category fit, review signals or the landing page itself.

Metric and control map

The findings refer to a small set of GMB Insights measures. Use this map to know what each one represents before you compare your own numbers:

  • Listing views — the denominator for the action rate; how often the listing was shown.
  • Website clicks — the most common customer action in the study and the metric used in the photo comparison.
  • Calls — phone taps from the listing; car dealerships led here.
  • Direction requests — map/directions taps; dealerships ranked second.
  • Photos — image count attached to the listing; average was 11.
  • Discovery vs brand searches — the split of views by query type; 84% discovery in the study.

These are descriptive measures, not platform configuration fields. Record them consistently so month-to-month comparisons are meaningful.

How to use these figures responsibly

The study is a dated snapshot: listings were analysed over a period ending before 2019, and the platform has changed since. The percentages and ratios are best used as a historical reference and a prompt for your own measurement, not as a claim about current performance. Where the study itself flagged a limit — notably the correlation-only nature of the photo finding — that limit still applies.

Frequently asked questions

Does the study prove that adding photos causes more clicks?

No. The analysis found a clear correlation between photo count and customer actions, but explicitly could not establish causation — businesses with many photos may be better at local marketing overall. Adding high-quality images is still a reasonable, low-risk optimisation.

What counts as a discovery search?

A non-branded query where the searcher is not looking for a specific business by name — examples given in the study include "bars near me", "dentist in brooklyn" and "greek restaurant". In the study, 84% of listing views came from this type of search.

Are the 2017 and 2018 action-rate figures still current?

They are a historical comparison from the study period. Use them to understand the direction of travel, then measure your own views-to-action rate against your own baseline rather than assuming the 2018 figure applies today.

Why were car dealerships the standout industry?

Dealership listings ranked third for average photo count, second for direction requests and first for calls and website clicks. The study's interpretation was that dealership listings tended to be well optimised and the sector had embraced the platform early — despite car buying being a high-ticket, low-frequency purchase.