Google Sheets URL Monitor Actor
Pricing
from $30.00 / 1,000 results
Google Sheets URL Monitor turns a plain spreadsheets into a **no-code web monitoring dashboard**. It reads the URLs listed in your sheet's `URL` column, visits each page in a real browser, extracts the values you care about (prices, titles), and writes the fresh values straight back into the sheet.
Full URL of the Google Sheet to monitor. Row 1 must be a header row containing a column named URL. Every other non-empty header is either a CSS selector whose extracted text is written into that column, or a plain field name (e.g. Price) that is extracted by AI when an OpenAI API key is provided.
Contents of a Google Cloud service account JSON key file with the Sheets API enabled. Share the spreadsheet with the service account's client_email as an Editor. If omitted, the Actor runs in read-only mode: it reads a link-shared sheet via CSV export and stores results only in the dataset.
Enables AI extraction for columns whose header is a plain field name (e.g. Price) instead of a CSS selector. The rendered page text is sent to OpenAI, which returns the requested fields. If omitted, the OPENAI_API_KEY environment variable is used instead (set it as a secret environment variable in the Actor settings, or in a local .env file). Without either, such columns are left untouched.
OpenAI model used for AI extraction of plain-name columns.
Optional instructions guiding how the AI fills the plain-name columns, e.g. Price must be a plain number without currency symbols. Availability is either 'In stock' or 'Out of stock'. Answer in English.
Name of the header column that receives the timestamp of the last successful update.
How long to wait for each CSS selector to appear on the page before treating it as missing. Selectors are awaited concurrently, so this bounds the extra wait per page.
Maximum number of pages processed in parallel. Keep low (1-5) since each page runs in a full browser.
How many times a blocked or failed page is retried before its row is skipped. Sites like eBay often return 403 a few times before letting a request through - raise this if such rows keep failing.
Minimum delay between two requests to the same website. Spacing out hits to bot-protected sites (eBay, Amazon) reduces 403 blocks.
