The Scenario
You are a retail data analyst. Your team just licensed a foot traffic dataset from a third-party provider, and the join key is Placekey. The problem: your internal store location list — 2,000 rows in a Google Sheet, columns A through D holding street address, city, state, and postal code — has no Placekey column at all.
The data partnership goes live in four days. Your manager has already announced the analysis in the quarterly roadmap. The Placekey column needs to exist.
The bad version:
- Open the Placekey developer docs, generate an API key, and figure out the bulk submission format for a two-thousand-row CSV.
- Write (or ask someone to write) a script that reads the sheet, batches the calls in groups of 100, handles retries on rate-limit responses, and writes the results back without scrambling the row order.
- Wait for the script to run, review the output for nulls, fix the rows where the address was incomplete, re-run for those rows, paste the final output into the sheet.
The analysis deadline is not a suggestion. Every hour you spend managing that pipeline is an hour you're not building the model the foot traffic data was supposed to power.
The Easy Way: One Prompt in SheetXAI
SheetXAI is an AI agent that runs inside your Google Sheet. It reads your columns, understands what you're looking at, and through its built-in Placekey integration it submits your address data to the API and writes the results back directly. No script, no export, no import.
Here is the prompt:
For every row, submit the address in columns A–D to the Placekey bulk API in batches of 100 and write the returned Placekey to column E
What You Get
- Column E fills with Placekey IDs — one per row — formatted as the standard '@here-222@5vg-82n-kzz' string Placekey returns.
- Rows where Placekey can't match the address get a blank in column E, not a silent skip, so you can filter for blanks and review the unmatched addresses.
- Batching is handled internally — SheetXAI submits 100 rows at a time and reassembles the results in original row order.
- The process runs to completion and reports how many rows were matched, how many returned no result.
What If the Data Is Not Quite Ready
The postal code column has mixed formats — some five-digit, some ZIP+4
Submit all rows to Placekey using street address (A), city (B), state (C), and zip (D), normalizing any ZIP+4 values in column D to five digits before submission, and write the Placekey to column E
Some rows have a blank city or state — submit what's complete and skip the rest
For each row where columns A, B, C, and D are all non-empty, submit the address to Placekey bulk API in batches of 100 and write the result to column E — leave column E blank for any row with a missing address field and add a note 'INCOMPLETE' in column F
The location list spans multiple tabs and needs to be processed together
Collect all rows from the Stores tab and the Franchise Locations tab, submit all addresses to Placekey in batches of 100, and write the Placekey back to column E in each tab, matching by row number
The full cleanup-and-enrich pipeline in one shot
Normalize state abbreviations in column C to two-letter codes, strip any extra whitespace from columns A–D, then submit all complete rows to Placekey bulk API in batches of 100 and write the returned Placekey to column E — flag any rows that return no match with 'NO MATCH' in column F
The most durable approach: ask for the normalization and the API call in the same prompt so the data going to Placekey is already clean.
Try It
Open a Google Sheet with your location list and get the 7-day free trial of SheetXAI, then paste the prompt above into the sidebar. Also worth checking out: enriching POIs with coordinates and the full Placekey overview.