ScrapeField
Owners and agencies watch reviews because a bad one answered in a day is half as bad as one left for a month. This guide builds a Make scenario that runs every morning, fetches the newest reviews of your places, and adds the new ones to a Google Sheet, where they can be answered, counted or sent on.
The modules
- Google Sheets, Search Rows, on a tab that lists your places, one
place_ida row. - HTTP, Make a request, once for each place.
- Iterator, over the reviews in the answer.
- Google Sheets, Add a Row, on a second tab, for each review that is new.
Schedule the scenario once a day.
The HTTP module
| Setting | Value |
|---|---|
| URL | https://api.scrapefield.com/v1/google-maps/reviews |
| Method | GET |
| Query parameters | place_id = the place_id from the first module, and sort = newest |
| Authentication type | API key: in its keychain, the header Authorization with the value Bearer sf_live_… |
| Parse response | Yes |
Make recommends the credentials field over a plain header, and it is the better place: the key is not in the scenario, so sharing or exporting the scenario does not share the key.
Only the new reviews
The answer’s data holds the newest reviews, up to 8 on the first page. Point the Iterator at data. Before Add a Row, add a filter that keeps a review only when its review_id is not already on the second tab. A Search Rows on that tab, by review_id, does it. Google’s own id never changes, so a review is never added twice.
Map the columns you want: rating, text, publish_time, author.name, and owner_response.text to see which ones have been answered. A rating with no words has text null, an empty cell in the sheet, never a made-up string. Every field is in the reviews reference.
What it costs
One page a place a day: 6 credits, so ten places for a month is 1800 credits, which is $1.40 on the smallest pack. A place that gets more than 8 reviews in a day needs the next page too, and so does the first run. The module’s Pagination section follows the pages: choose the token or cursor type, set Next page token/cursor path to meta.next_cursor and Page token/cursor parameter name to cursor, and tell it the items are in data. Each page is 6 credits, so for a place with thousands of reviews, stop at the first review you already have rather than following every page. Every list pages the same way (how lists work), and every review of a place shows the same in code.