I have a legacy database with "text" addresses and places. Examples:
1231 Main Street, Boston, MA
Berkeley, CA
...
...
I would like to transition to Google Places API and replace those with a place_id. In order to do that I would have to run a one-time process on our backend to submit those addresses to google and retain the place_id I get back.
The problem is that their agreement prevents "pre-fetching". Being blocked by Google would be a death sentence for my business. I would avoid that at all costs.
What can I do in order to migrate my tens of thousands of "text addresses" to place_id without being accused of "pre-fetching"?
Now I do understand place_id is exempt from caching restrictions, but when I use the geocode API to query an address, I get back a whole structure containing much more than just the place_id. I get the formatted address, the breakdown to address components, and long/lat. My fear is that my attempt to migrate to Places API by submitting tens of thousands of requests from my backend cluster will seem like a scraping attempt. Is there way to restrict the output to contain JUST the place_id - maybe that would help avoid any compliance issues?
On the Place IDs page it says:
Place IDs are exempt from the caching restrictions stated in Section 3.2.3(a) of the Google Maps Platform Terms of Service. You can therefore store place ID values for later use.
So, no need to worry about breaking the Code of Conduct.