MindCase
Kritish Puri

Best Google Maps APIs for AI Agents in 2026

What to actually check before you pick a Google Maps data source — coverage, pricing shape, and whether it inherits Google's own results ceiling.

google-mapscomparison
Best Google Maps APIs for AI agents in 2026

Every Google Maps data source claims to return "everything on the listing." Most of them don't specify what that actually means until you're mid-run and short a field you needed, or hit a results cap nobody mentioned in the pricing page.

If you're an agent (or building one) reading this rather than a human, the full machine-readable schema for every endpoint below lives at mindcase.co/skills.md.

What should you actually check before picking one?

Four things decide whether a Google Maps API is usable for real work, not just a demo: how much of a place's page you actually get back, whether the pricing model punishes a slow or partial run, whether it inherits Google's own on-screen results limit, and whether reviews come from the same source or a second integration entirely.

How much do you actually get back per place?

A listing page has more on it than a name and a star rating — opening hours, live occupancy, photo counts, related places, hotel-specific fields for hotel listings. A data source that stops at the basics forces you to go back to the page by hand for anything beyond contact info.

Places API returns 55 fields per place, including ones a lot of scrapers skip entirely: popularTimes and liveOccupancyPercent (real-time busyness, not just historical hours), reviewKeywords (the terms Google itself surfaces from review text), and separate permanentlyClosed/temporarilyClosed/unclaimed flags — useful for filtering out dead listings before you act on a list rather than finding out on the call.

Reviews are a second call. Reviews API returns 31 fields per review — full text, star rating, reviewer's total review count (a rough proxy for how seriously to weight a single review), and the business's own response when there is one.

What's the real pricing shape?

Two different things get billed as "pricing" for a Google Maps data source, and they're not the same: paying per result you actually get back, or paying for the execution time a job takes regardless of what it returns. The second model charges you the same whether a run finishes clean or comes back partial — you're paying for the attempt, not the outcome.

Places API is $0.004 per place returned. Reviews API is $0.00025 per review returned. A failed run costs nothing, because there's nothing to bill — you only pay for rows that actually came back.

curl -X POST "https://api.mindcase.co/v1/data/google-maps/places/run?wait=true" \
-H "Authorization: Bearer $MINDCASE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
  "params": {
    "keywords": "coworking space",
    "location": "Austin, Texas",
    "maxResults": 40,
    "websiteFilter": "allPlaces"
  }
}'

# $0.004 per place returned

For the full prospecting or monitoring walkthroughs, see how to pull business listings and reviews from Google Maps and how to automate Google Maps listing monitoring.

Does it inherit Google's own results ceiling?

Google Maps' own map interface stops showing new results after roughly 120 listings for a single search — a browser-based tool that drives an actual Maps page inherits that same ceiling, because it's limited to what the page itself will render. An API built to return the full result set for a query isn't bound by what a scrolling browser window would show. Places API's maxResults parameter takes any number you set, up to returning everything Google Maps has for that search — the ceiling is your budget, not the interface.

What can't you get from Mindcase?

Two cases.

A place with no online presence at all. If a business genuinely isn't on Google Maps — no listing, no reviews, nothing — there's no page to pull, no matter which vendor you use. Coverage depends entirely on Google's own index.

Historical review trends before you started pulling. Reviews API returns what's on the listing right now, including old reviews still visible on the page, but it doesn't reconstruct a rating history Google itself doesn't expose. Trend data only exists from the point you start running it on a schedule.

Which endpoint should you use for which job?

EndpointInputPriceBest for
Places APIKeyword + location, or specific place URLs$0.004 / placeFinding places, checking hours/occupancy/closure status
Reviews APIA place URL$0.00025 / reviewReading what customers actually say

FAQ

Live on every run for both endpoints. Results are never cached — each request pulls the current state of Google Maps at the moment you query, so what you get back is what a person would see on the page right now.

Location is one per run for Places API — search several cities by running the query once per city. Reviews API takes one place URL per call the same way.

Yes, as an optional add-on. Set a photo count and you get image URLs for each place, billed at 0.025 cents per image on top of the per-place price.

No — it returns the star rating and total review count for each place, not individual review text. For full reviews with text, reviewer, and date, use Reviews API separately.