Google Maps returns about 200 results, and no query gets you more
The cap is per query, not per page, so the answer is geographic rather than paginated. What saturation means, why the threshold is 190, and how deep the splitting goes.
Google caps any single Maps query at about 200 results. That is a property of the query rather than of a page, so no amount of paging, offsetting or re-asking gets past it — the only way to see more businesses is to ask about smaller areas. This Actor treats a tile returning at or above 190 results as saturated and splits it into four, repeating until every tile comes back under the threshold. The threshold sits at 190 rather than 200 because the same query does not return the same count twice, and a dense 1-degree cell over a capital city was still splitting at depth 8.
Key points
The cap applies to the query, so pagination cannot reach past it.
A tile at or above 190 results is treated as saturated and split into four.
190 rather than 200 leaves margin, because the same query does not return the same count twice.
The stopping rule is saturation, not depth: `maxDepth` is a safety guard, and reaching it means the census is truncated.
A dense 1-degree cell over a capital city was still splitting at depth 8, which is why the guard defaults there rather than lower.
Everything about this Actor's design follows from one number, so it is worth being precise about what that number applies to.
The cap is per query
Google returns about 200 results for a Maps query and stops. It is not a page size with more pages behind it, and it is not a rate limit that lifts. Requesting the same area again returns the same bounded set, so no client-side strategy — paging, offsetting, retrying, sorting differently — reaches business 201.
What does change the answer is changing the question. A smaller area has fewer businesses in it, and if it has fewer than the cap, you see all of them.
Saturation, and the number that defines it
A tile that comes back at or above the saturation threshold is assumed to be hiding results behind the cap, so it is split into four and each quarter is retried. A tile that comes back under the threshold is assumed complete.
Why 190 and not 200
Because the same query does not return the same count twice. A threshold set exactly at the observed ceiling would sometimes read a full tile as complete, and a tile wrongly marked complete is invisible — it produces missing businesses rather than an error. 190 leaves margin for that variation, which costs some unnecessary splits and buys correctness.
Depth is a guard, not the stopping rule
maxDepth defaults to 8 and is a safety guard. The actual stopping rule is saturation: splitting continues until tiles come back under the threshold. If a run reaches the depth limit, its census is truncated, and both the log and the run summary say so rather than letting the result pass as complete.
The default sits at 8 because a dense 1-degree cell over a capital city was still splitting at that depth — so a lower guard would silently truncate exactly the areas with the most businesses in them.
About 200 per query. It is a limit on the query rather than on a page, so paging, offsetting or repeating the request does not return more — only asking about a smaller area does.
▸What counts as a saturated tile?
One returning at or above the saturation threshold, which defaults to 190. Saturated tiles are split into four and retried until every tile returns under the threshold.
▸Why is the threshold 190 rather than 200?
Because the same query does not return the same count twice, so a threshold set exactly at the ceiling would sometimes fail to notice a full tile. 190 leaves margin for that variation.
▸How deep does the splitting go?
As deep as saturation requires; `maxDepth` is a safety guard defaulting to 8, and a run that reaches it has a truncated census, which both the log and the run summary state. A dense cell over a capital city was still splitting at depth 8.
Sources
Every URL below was requested and returned a page on the date shown.
Google's own API is the right tool for looking up a place. It is a different proposition when the question is every business in a region, and the difference is structural rather than about price alone.
The first pass extracts no businesses at all. It tells you which parts of an area are dense, which belong to a neighbouring country and which are empty — before you spend anything.
