Findings · 2026-09-25 · By VSNARY | Emmanuel Orta · 0 views

Google Places API rankings are not what Google Maps shows

Same query, four Denver-area cities, one day. The API and Maps agreed on who competes, disagreed on the order, and the API could not find two businesses Maps ranks.

If you track local rankings with Google’s Places API, you are not tracking Google Maps. We asked both the same question, “tree service in <city>, CO”, for four Denver-area cities on 25 September 2026. The two lists agreed on who exists. They did not agree on the order, and the API could not find two businesses that Maps ranks.

What we compared #

The Places API column is Text Search (New), the places:searchText endpoint, with rankPreference: RELEVANCE, a location bias of an 8 km circle around the city centre, and up to 20 results. The Google Maps column is the public Maps results page for the same query with the map centred on the same city, read by our own monitor through a headless browser, signed out, with sponsored results skipped. Maps was read at about 10 AM Mountain Time and the API later the same afternoon.

Same words, same city centre, same day. If the two sources ranked the same way, the lists would line up. They did not.

The top of Maps, and where the API put the same businesses #

Read the Denver row: the business Maps shows first is fifth in the API, and Maps’ third is thirteenth. In Aurora, Maps’ first result is seventh in the API. Only three of twelve top-three slots matched exactly (Arvada #1 and #3, Lakewood #3), and no city had the same top three in the same order. Every Maps top-three business did appear somewhere in the API’s list, so the API is a fair census of who competes. It is not a ranking of who wins.

The businesses the API could not find #

Two of the businesses our monitor tracks are service-area businesses: they serve customers at their homes and hide their street address on their Business Profiles. Here is what each source said about them.

The second row is the one to remember. We searched the API for the Arvada business’s exact name, as it appears on its profile, and got fourteen other tree companies and not the business itself. The profile is live and reviewed, and Maps ranks it on page one. Text Search does not return it. A tracker that finds businesses by searching for them would report this one as not ranking anywhere, when it is fifteenth in the list searchers actually see.

Neither business was missing from Google. Both were missing from one way of asking Google.

Why this matters if you sell or buy rank tracking #

• A rank from the API is a rank in a different list. A tool that reports “#1 in Aurora” from Text Search can be describing a business that searchers see second, or seventh.
• Service-area businesses can vanish. If your tracker searches by city or name, a hidden-address profile can read as “not ranking” while it sits on page one of Maps.
• Moves are only comparable within one source. We compare Maps readings with Maps readings and label any API fallback as such. Mixing them manufactures rank changes that never happened.

What we changed in our own monitor #

Our local-rank monitor now reads positions from Google Maps itself, rendered from each city’s centre, and keeps the Places API only as a fallback. Any position that came from the fallback is marked “[API reading, Maps rescan queued]” so it is never mistaken for a Maps position. The businesses we track are matched by place ID, not by name, so a profile the search cannot find is still recognised when Maps shows it. The monitor also reports each city in five buckets — top 3, 4–10, 11–20, not in the top 20, and queued — so a business at #12 no longer disappears into “not top 10”.

What the Places API is still good for #

None of this makes the API useless. It returned every business that Maps put in a top three, so it is a sound way to build a list of competitors in an area. It can return structured fields — place ID, categories, website, rating and review count — that are tedious to read from a rendered page. And it is the documented, supported way to get that data. Use it to learn who is in the market. Use Maps to learn who is winning it.

How to check your own tracker in ten minutes #

1. Pick one query and one city your tool reports on.
2. Open Google Maps signed out, search the same query, and write down the first ten names.
3. Compare them with your tool’s list. If the top three differ, ask the vendor where its positions come from.
4. If you run a service-area business, search the API or your tool for your exact business name. If you are missing, your tool is not measuring you at all.

What this does not show #

• Four cities, one trade, one day. It shows the two orders differ; it does not measure by how much across the country.
• Google does not document how Maps ranks results or how Text Search’s relevance ranking is built, so we make no claim about why they differ.
• Maps varies with location, device and personalisation. Our reading is one signed-out vantage near each city centre.

Related: what Google’s Maps provider list contains and the business-data suppliers it names.

Questions this post answers

Does the Google Places API return the same ranking as Google Maps?

Not in our test. Across four cities only three of twelve Maps top-three positions matched the API exactly, and Maps' first result was fifth and seventh in the API in two cities.

Why is my business missing from Places API search results?

In our test, two service-area businesses with a hidden address were not returned by Text Search for their city, and one not even for its exact name, while Maps ranked them #12 and #15.

Can I use the Places API for local rank tracking?

It is a fair list of who competes, but not a reliable order. Read positions from Google Maps itself, and never compare an API position with a Maps position.

How do I check where my local rank tracker gets its data?

Search the same query in Google Maps signed out, compare the top ten with your tool, and search your tool for your exact business name. Ask the vendor if either differs.

Does this prove how Google ranks Maps results?

No. Google does not document either ranking. It shows the two orders differ on the same day for the same query.

Related findings

Comments

Comments are read before they appear. Nothing is published automatically, and no account is needed.

Writing about this? Facts, live figures and marks — every number on that page is dated and traceable to a scan.

All findings · The dataset · How the dataset works