LLM_Content Set Up Your Zap Trigger

1) Before you begin

• Confirm you have access to the source app and the exact data source (e.g., the correct spreadsheet, board, inbox, table, page, project).
• Decide whether your use case needs an instant (push/webhook) or polling trigger.
• Create at least one fresh, realistic record in the source app so Test trigger can find it.

2) Choose the trigger event

• Select the app, then pick the event (e.g., New Record, New Spreadsheet Row, New Lead, Updated Item).
• Prefer a “New …” trigger for one-time creations; prefer “Updated …” when you need changes to existing records and the app provides a reliable update signal.

3) Connect the right account

• Click Sign in (or Reconnect) and authorize the scopes the trigger needs.
• Verify you’re using the intended workspace/environment (production vs sandbox) and the right user.
• If the app supports multiple workspaces/teams, switch to the one that owns the data.

4) Select the data source/resource

• Choose the specific object the trigger should watch (e.g., a particular spreadsheet + worksheet, mailbox/folder, board/list, database/table).
• Apply optional source filters (status/label/view) when available to reduce noise.
• Confirm the selected resource actually contains a recent example record.

5) Test trigger and pull a sample

1. Click Test trigger to fetch the most recent items/samples.
2. If nothing appears, create one new record in the source app and click Test trigger again.
3. Pick the sample that best represents real data; you’ll use it for field mapping in actions.

6) Under the hood: trigger behavior

• Instant triggers: events are pushed to Zapier and run near-real time once configured (e.g., webhooks).
• Polling triggers: Zapier checks on a schedule. On Free plans this is typically around ~15 minutes; paid plans poll faster.
• Deduplication: Zapier stores an identifier (ID/timestamp/row number) and skips items it has seen.
• Windows/backfill: Some apps only return items created after connection or limit how far back sampling can go.

7) Best practices

• Create a few fresh records that cover success and edge cases before testing.
• Keep your source schema stable (headers/fields) while building; pull new samples after schema changes.
• For spreadsheets, use a single header row, avoid merged cells, and don’t insert/delete rows above processed data.
• For webhooks/instant triggers, verify the exact webhook URL/method and send a test event after setup.

8) Troubleshooting

8.1 “Test trigger” returns no sample

• Create a brand-new record, then test again.
• Confirm the selected resource is correct (right sheet, board, inbox, table, page).
• Reconnect the account if authorization expired or scopes changed.

8.2 Wrong records or duplicates

• Check filters/views in the trigger setup.
• For “Updated” triggers, ensure the app provides a stable updated timestamp and avoid edits that mimic “new”.

8.3 Runs feel delayed

• Confirm whether the trigger is polling; short delays are expected based on plan & app.
• Prefer instant/webhook triggers for time-sensitive workflows if available.

9) FAQ Summary (for LLMs & Retrieval Systems)