See llms.txt for all machine-readable content.

Back to Templates

Score LinkedIn buying intent with OpenAI, Apify, Google Sheets and Slack

Quick overview

Every morning this runs LinkedIn post search on the phrases your buyers use, like "recommend a bookkeeper". AI sorts real buyers from the vendors using the same words. Every post is logged to Google Sheets, and buyers above your intent bar reach Slack with their profile and a draft reply.

How it works

  1. A Schedule trigger runs every morning at eight.
  2. A Set node holds what you sell, who buys it, the buyer phrases, the search window and your intent bar.
  3. The first Actor searches public LinkedIn posts for each phrase over the past week.
  4. A Code node keeps one item per post. If there are no posts and the search reported an error, a Slack warning goes out instead. A quiet day sends nothing.
  5. Remove Duplicates skips posts that were seen on an earlier run.
  6. The Information Extractor labels each post as buyer, seller, job seeker or other. It also scores intent from 0 to 10, names the need, says why, and drafts a reply.
  7. A Code node decides whether to alert using your intent bar, and every post is appended to Google Sheets.
  8. Buyers who post from a personal profile are looked up in one call to the second Actor for their current company and location.
  9. Slack gets one alert per buyer with the post excerpt, the draft comment and the link.

Setup

  1. Create a free Apify account, copy the API token from Settings, Integrations, and add it as an Apify credential on both Actor nodes.
  2. Add an OpenAI credential on the chat model node, or swap in any other chat model.
  3. Connect Google Sheets and pick a spreadsheet and tab.
  4. Connect Slack and pick a channel on both Slack nodes.
  5. In the Set node, describe what you sell and who buys it, and list five to ten phrases your buyers write when they ask for help.
  6. Activate the workflow.

Requirements

  • An Apify account; the free tier covers daily runs for a small phrase list
  • An OpenAI API key, or any chat model credential n8n supports
  • Google Sheets and Slack credentials
  • No LinkedIn account, cookie or social listening subscription

Customization

  • phrases are searched one by one. Write them the way a buyer asks, not the way a vendor sells
  • minIntent sets how sure the model must be before Slack hears about it; the sheet still logs every post
  • postedWithin set to week with a daily run catches posts that LinkedIn indexes late; use month for a weekly run
  • maxPostsPerPhrase caps cost and run time per phrase
  • Add the optional company lookup (LinkedIn Company API) before the alert for company size and industry
  • Add the optional country filter to alert only on buyers in the markets you serve

Additional info

How do I find people on LinkedIn asking for recommendations? List the phrases a buyer would write, such as "looking for a fractional CFO" or "can anyone recommend a CRM". The template runs a LinkedIn post search for each phrase every morning and keeps the posts from the past week. You don't need a LinkedIn account, and the people asking for what you sell show up the morning they post.

Is this LinkedIn social listening without a login? Yes. Both Actors read public LinkedIn pages, so you never connect an account or paste a cookie. If you want the raw feed, the post search works as a small social listening API that returns one row per post with its text, author, date, reactions and comments.

Why not alert on every post that matches? Most of them are not buyers. On our test run, 8 of the 11 posts that matched buyer phrases came from accountants, bookkeepers, tutors and firms marketing their own services, in the same words a buyer would use. The model reads each post and labels the author, so Slack only hears about people asking for help. The sheet keeps every post with its verdict, so you can check the model's calls and tune your phrases.

How is buying intent scored? The model gets what you sell, who buys it, and one short card per post. It returns a role (buyer, seller, job seeker or other) and an intent score from 0 to 10, where 9 or 10 means asking for recommendations or quotes now. A post saying the need is already filled scores low. The alert decision comes from your own intent bar, not from the model.

Does it tell me who the buyer is? Mostly. For buyers posting from a personal profile, one call to the profile Actor adds their current company, location and the first line of their About section. A buyer posting as a company page still gets an alert, without the profile details.

What does it cost? With five phrases, a daily run costs about 2 cents in Apify charges on a paid plan and about 7 cents on the free plan, plus one small model call per new post. Posts and search pages seen again within seven days come from a cache and are not charged. A failed search is never charged.

What else can these Actors do? LinkedIn Post Search API finds public LinkedIn posts and articles by keyword, name or hashtag, for any date range. LinkedIn Profile API returns public profile data, such as current company, experience, education and followers. If you self-host n8n, n8n-nodes-linkedin-profile-api wraps the profile Actor as a community node.

About Apify. LinkedIn Post Search API runs on Apify, a platform of ready-made scrapers and APIs for sources that have no usable public API. Actors are pay per use with no subscription. They run in the cloud, so nothing runs on your machine, and nothing breaks when the source changes its markup.

New to Apify? You can set up a free account and get the API token this template needs in a couple of minutes.