Telegram lead search bot that saves leads to Google Sheets

Telegram lead search bot that saves leads to Google Sheets

An n8n workflow behind a Telegram bot. You message the bot a job title and a state in plain words. A model pulls the
two out, an Apify Actor searches for people who match, and the workflow checks on the search every few minutes. When it
finishes, the leads with an email are added to a Google Sheet, and the bot replies with how many it saved.

Read "The Actor has changed" below before you build on it: the Actor named in the file now takes a different input.

What is inside

• workflow.json: the workflow as n8n exports it, nineteen nodes and no sticky notes. It holds no credential.
• README.md: these notes.

What we changed from the original copy

• The Apify headers in runActor and getDatasetItems held a placeholder, Bearer <token>. They now say
  Bearer <YOUR_APIFY_TOKEN>.
• The author's sheet and its tab id are gone. Google Sheets asks for YOUR_SHEET_ID and finds the tab by the name
  Leads.

Before you start

• An n8n instance that Telegram can reach over HTTPS. n8n Cloud is, and a server of your own needs a public address.
• A Telegram bot and its token, from BotFather.
• An OpenAI API key that can use gpt-4.1-mini.
• An Apify account and its API token, with credit, and a lead search Actor that takes job titles and states (below).
• A Google account, and a Google Sheet for the leads.

The Actor has changed

runActor starts the Actor GlxYrQp6f3YAzH2W2 with a job title, a state and a few email settings. On 2 November 2026,
Apify's listing for that ID named it Email & Phone Finder, from the account peakydev. Its input page listed one field, a
CSV or Excel file of people, and its pricing history said the Actor was turning into a LinkedIn profile scraper. The
search this bot sends no longer fits it.

To replace it, pick an Actor that searches people by job title and state. The Leads Finder Actor from the account
boneswill, for one, lists personTitle and personState on its input page. Then:

1. Put the new Actor's ID in the address of runActor, and write the body its input page asks for.
2. Change the address of getRun to https://api.apify.com/v2/actor-runs/{{ $('runActor').item.json.data.id }}, which
   works for any Actor.
3. Run one search and match setValues to the fields the new Actor returns. Check If too (How a run goes, step 6).

Set it up

1. In n8n, create a workflow, open its menu and choose Import from File. Pick workflow.json.
2. Make a Google Sheet with a tab named Leads and these headers in row 1: personName, title, email, linkedIn,
   person_location and organization_website. Copy the sheet's ID: the long part of its address between /d/ and
   /edit.
3. Add a Telegram credential with your bot's token. Select it in Telegram Trigger and in the three nodes that send
   messages.
4. Add an OpenAI credential and select it in OpenAI Chat Model.
5. Add a Google Sheets OAuth2 credential. In Google Sheets, select it and replace YOUR_SHEET_ID with the ID.
6. Replace the Actor (above). In runActor and getDatasetItems, replace <YOUR_APIFY_TOKEN> in the Authorization
   header with your token. The brackets go too, so the header reads Bearer, a space and the token. Give getRun the
   same header, since it has none in the file.
7. If no one else should use the bot, add an If node after Telegram Trigger that lets through only your own user ID,
   {{ $json.message.from.id }}. The trigger's own Restrict to User IDs option shows in this node, but n8n's code only
   applies it from version 1.2 of the node, and this one is 1.1.
8. Save and press Publish. Then send the bot a message such as "marketing directors in Texas".

How a run goes

1. Telegram Trigger starts a run for every message the bot receives.
2. Query Fixer has gpt-4.1-mini pull out two fields: jobTitle, and personState when the message names a state.
   Its prompt tells it to write both in a standard form.
3. runActor starts the Actor with the title, the state, verified emails only, leads without an email skipped and up to
   1000 results. It does not wait for the search.
4. Wait pauses two minutes, then getRun asks Apify for the run's status. Switch sends FAILED to
   Send a failed message, SUCCEEDED to getDatasetItems, and RUNNING to Wait1, which pauses three minutes and
   asks again.
5. getDatasetItems fetches the results, and Aggregate gathers them into one list called data.
6. If checks whether the second item in that list has an id. When it does, the workflow takes the results for an
   error notice, and Send a error message replies with the first sentence of that item's fullName. That test fits how
   this Actor shaped its results when the workflow was made.
7. Otherwise Split Out turns the list back into one item for each lead. setValues keeps six fields and joins the
   city, state and country into person_location. Next, Filter drops leads without an email and Remove Duplicates
   drops a repeated email.
8. Google Sheets appends each lead as a row, and Send a success message replies once with the number of rows, the job
   title and the place.

Change the replies

• The three messages: the text of Send a success message, Send a failed message and Send a error message. The
  success message ends by thanking the user for using LeadGenBot, so put your bot's name there.
• The wait: Wait before the first check and Wait1 between checks.
• The search: the body of runActor, and the two fields Query Fixer extracts.

Good to know

• Anyone who finds your bot can start a search that spends your Apify and OpenAI credit. Step 7 of the setup closes that.
• Switch has routes for three statuses only. A run that is still queued at the first check, or one that times out or is
  aborted, ends the workflow with no reply.
• The extractor reads a state, not a city or a country. A message without a state sends an empty state to the Actor.
• Remove Duplicates only compares the leads of one search, so a second search can add people already in the sheet.
• The success message reports the rows saved in this run, after the leads without an email are dropped.
• Each message goes to OpenAI, and the job title and state go to Apify. The sheet holds names, emails and LinkedIn
  addresses. Check the rules for contacting people where you and they are before you write to anyone.
• The Apify token sits in headers, so take it out before you share the workflow.