Articles on: Spreadsheet

Connect a real-time data source

Lido tables can pull rows from a CRM, database, generic API, or file — and refresh them automatically. This is how you keep a sheet in sync with the system of record without manual exports. This article covers the setup wizard and what to do once a source is connected.



When to connect a real-time source


Connect when:


  • The data lives in another system and you don't want to copy-paste it.
  • You want the spreadsheet to reflect the latest values without manual refreshes.
  • You want to combine external data with computed columns, AI calls, or notification actions.


Don't connect when:


  • The data is a one-time import that won't change. Just paste it in or use Create Table from a range.
  • You only need a snapshot. Export from the source system and upload the file.
  • The volume is huge (millions of rows). Connect a smaller filtered view instead.



Step-by-step: connect a source


  1. Open File → Connect real-time data in the spreadsheet menu.
  2. Pick a source type from the wizard. Lido supports common CRMs (Salesforce, HubSpot, Pipedrive), databases (Postgres, MySQL, MSSQL, BigQuery, Snowflake), generic APIs, file connections (Google Sheets, S3), and more. The exact list depends on your plan.
  3. Authenticate. For OAuth-based sources (CRMs, Google Sheets), you'll click through the source's OAuth flow. For databases, you'll enter connection credentials directly. For generic APIs, you'll provide an endpoint URL and auth headers.
  4. Pick the records you want. Each source has its own selection step — pick a Salesforce object, a Postgres table, an HTTP endpoint, etc.
  5. (Optional) Apply filters at the source level if available. Pulling a filtered view is faster and uses less of your plan than pulling everything and filtering inside the sheet.
  6. Configure the refresh schedule — manual, every X minutes, daily, etc.
  7. Insert the table. Lido creates a named table with one source column per field returned by the source.


The table name and column names default to what the source provides. Rename them if you want — the table reference syntax (MyTable[@Column]) tracks renames.



What happens after the source is connected


  • Source columns refresh on the schedule you set, or when you manually click Refresh.
  • You can add other column types alongside source columns: computed (formulas), linked (per-row data), or plain (manual entry). They live next to the source columns in the same table.
  • You can reference the table in formulas anywhere in the workbook: =SUM(Salesforce Opps[Amount]).
  • Other Lido features can use the table: workflows (Get Table node), AI formulas (GPT(..., Salesforce Opps[@Notes], ...)), and exports.



Step-by-step: enrich source data with AI


Common pattern: pull records from a CRM, then add an AI-generated summary or classification column.


  1. Connect a Salesforce source — table called Opportunities with columns Name, Stage, Amount, Notes.
  2. Add a computed column AI Summary:
   =GPT("openai-cred",
        "Summarize this opportunity in one sentence: " & Opportunities[@Notes],
        Opportunities[@AI Summary])
  1. Add a computed column Risk that classifies risk level:
   =GPT("openai-cred",
        "Rate the risk of this opportunity 1-5 based on: " & Opportunities[@Notes],
        Opportunities[@Risk])
  1. Source columns refresh from Salesforce; computed columns recompute when source values change.



Step-by-step: keep extracted document data alongside source records


Pattern: invoices arrive as PDFs, but you also have the vendor in a CRM. You want both.


  1. Connect a CRM source for vendors.
  2. Build a separate extractor worksheet for invoice fields.
  3. Use a workflow: Google Drive Trigger → Data Extractor (invoice config) → Smart Lookup (match Vendor Name to the CRM table) → Insert Rows.


The result: an invoice tracker that links extracted invoice data to the canonical vendor record from the CRM.



Tips


  • Filter at the source. Pulling 100,000 rows when you only need 500 wastes refresh time and may exhaust connector limits.
  • Refresh as rarely as you can stand. A 5-minute refresh on a CRM with thousands of rows is wasteful if the data only changes a few times a day.
  • Use source-side identifiers as your row IDs. When you add linked columns or merge with other sources, you'll want a stable unique key — usually the source system's record ID.
  • Combine sources via Smart Lookup or formulas. Lido tables don't have a SQL JOIN, but the Smart Lookup node and lookup formulas (VLOOKUP, XLOOKUP, MATCH) get you most of the way.



Common mistakes


  • Refreshing on a schedule that's faster than the source can handle. Some APIs rate-limit aggressively. Watch for refresh errors.
  • Editing source column values manually. They'll be overwritten on the next refresh. Use a separate plain or linked column.
  • Connecting the wrong account. OAuth flows can grab whatever account is currently logged into the browser. Verify the connected account matches what you intended.
  • Pulling massive tables when you only need a slice. Filter at the source if the connector supports it.
  • Forgetting to handle source outages. If the source is unreachable at refresh time, the table shows stale data. Subscribe to error notifications in workspace settings.




  • Tables: how columns work
  • AI columns and formulas
  • Build your first workflow
  • Connect Google Drive
  • Connect OneDrive
  • Lido API: quickstart and authentication

Updated on: 16/04/2026

Was this article helpful?

Share your feedback

Cancel

Thank you!