ChatGPT for Sales: Add Your LinkedIn Account to Call Prep and Follow-Up
ChatGPT for sales usually starts with a prompt list: draft the cold email, summarize the account, rehearse the objection. This guide covers one call and the part of it that lives on LinkedIn – who the attendees are now, who you already know at their company, and what you last said to each other – and how ChatGPT can work from your own LinkedIn account before the call and after it.
The short version. ChatGPT already drafts emails and researches companies, and through OpenAI's Sales plugin it works from your CRM, call recordings, Slack and Teams. LinkedIn's own plugin in ChatGPT looks up who someone is. Connect your own LinkedIn account through Linked API and ChatGPT also sees what only your account knows: the attendees' recent posts, who you know at their company, and your conversation with each of them. Before the call it writes the brief; after it, it drafts the reply and a connection note, and nothing is sent until you approve. You need a Linked API account with your LinkedIn account connected, added to ChatGPT as a custom connection – OpenAI offers that to Business and Enterprise/Edu workspaces on the web.
Before the call: a LinkedIn brief in ChatGPT
One ask, one brief – the setup is below:
Used Linked API: fetch_person, check_connection_status, fetch_company, sync_conversation, get_inbox
- experience, and posts and comments from the last 90 days
- connected to Priya; not connected to Tom, no invitation pending
- employees with "sales", then "revenue operations", in their position, at 1st or 2nd degree to you
- your message history with each attendee, loaded once
- the conversation with each attendee, by their profile URL
Brief ready. You're connected to Priya and last wrote to each other in September; Tom is a 2nd-degree connection you haven't messaged. At Halden, two people in sales roles are 1st-degree connections and six people in sales or RevOps roles are 2nd-degree – people to look into, not a ready introduction path. The three openers: Tom's posts on a monthly forecast, Priya's comment on CRM data hygiene, and the case study you sent her. Nothing has been sent.
| Attendee | Role and time in it | Recent topic they posted about | Connected? | Last exchange |
|---|---|---|---|---|
| Tom Becker | VP Sales, 1 year 7 months | Three posts on moving his team from a quarterly to a monthly forecast | No – 2nd degree | None |
| Priya Shah | Head of Revenue Operations, 3 years 2 months | A comment on a thread about CRM data hygiene | Yes – 1st degree | In September she asked for a case study on routing inbound leads; you sent it, no reply since |
nothing is sent from your account – the follow-up after the call waits for your approval
If more people sit behind the decision than joined the call, mapping the buying committee at the account is the next step.
What ChatGPT already does for sales – and what your LinkedIn account adds
Per OpenAI's Sales plugin page, the plugin is built to "Prepare account work and customer follow-up". The apps that page lists are Slack, Teams, Zoom, Granola, Fireflies, Otter.ai, Salesforce and HubSpot, plus "19 more". LinkedIn has its own plugin in ChatGPT, and its listing reads: "Look up any professional on LinkedIn without leaving ChatGPT. See their title, company, location and follower count", with "Prepare an attendee brief" among its example prompts.
Between them, those cover your CRM, your calls and who someone is. What neither describes is your own LinkedIn account – your network at the account and your conversations:
| Moment | OpenAI's Sales plugin and your apps | LinkedIn's plugin in ChatGPT | Your own LinkedIn account through Linked API |
|---|---|---|---|
| The account | Account research from your CRM and connected apps | Not in its listing | Who works there in the roles you name, at 1st or 2nd degree to you |
| The person | CRM records and call notes | Title, company, location and follower count, with a link to the profile | Their experience, recent posts and comments, and whether you're connected |
| The last exchange | Chat and call recordings | Not in its listing | Your LinkedIn conversation with them |
| After the call | Customer follow-up drafted from your apps | A link to the profile, to connect yourself | A reply in your LinkedIn conversation and a connection note, sent only when you approve |
What you need first
You need a Linked API account with your LinkedIn account connected to it, added to ChatGPT as a custom connection. Linked API works through a dedicated cloud browser on your own account, at a human pace, and never stores your password. The setup steps for ChatGPT are in the installation guide, and what the connection can do is covered in the overview.
- Your ChatGPT plan. Per OpenAI's help, "Custom MCP server connections and full MCP support are available for ChatGPT Business and Enterprise/Edu customers on ChatGPT web". On Business, "Only admins and owners can create and publish plugins for custom MCP servers"; on Enterprise/Edu, admins can let selected members connect one.
- On Pro. OpenAI allows Pro read and fetch tools only, and since version 2.3.24 Linked API marks which of its tools only read. By those two rules, Pro can look people, companies and posts up and read messages already captured, but can't load missing conversation history or send anything – both runs on this page need Business, Enterprise or Edu.
- Inbox monitoring, if you want it. Ask ChatGPT once to turn on inbox monitoring for your account if you want it to see new messages across your whole inbox. Reading one person's conversation works without it.
ChatGPT is one way in. The same account also works from Claude, Cursor or Claude Code through the MCP server, the agent-friendly CLI or a ready-made skill – the LinkedIn AI agent guide compares the routes.
After the call: the follow-up, approved
Paste your notes and ask for the drafts:
Used Linked API: get_inbox, check_connection_status
- your conversation with Priya, loaded before the call
- Tom – still not connected, no invitation pending
Two drafts, written from your notes and the brief. Nothing has been sent.
| To | Where it goes | Draft |
|---|---|---|
| Priya Shah | Your existing LinkedIn conversation | "Thanks for today, Priya. As agreed, I'll send the routing pilot plan by Tuesday, and we'll go through it with Tom next week." |
| Tom Becker | Connection request with a note (117 characters) | "Hi Tom – thanks for joining today. Your posts on monthly forecasting came up twice; glad to connect before the pilot." |
nothing goes out until you approve each draft – ChatGPT may also ask you to confirm before it sends
Per LinkedIn's help, a note can be up to 200 characters, and without Premium you can add one to up to three connection requests per month – see what to write in a connection request.
Did they reply? Once you've loaded a conversation, Linked API keeps adding its new messages for a while (30 days by default). Before you write again, ask ChatGPT to refresh the conversation and read it. An empty result means nothing new has been captured yet, not that they haven't replied.
ChatGPT's own approvals. ChatGPT has its own setting for how plugin actions are approved. Per OpenAI's help on app permissions, with "Always ask", "ChatGPT asks before reading app information or making changes", and "Sending an email, message, comment, or invitation" is one of its examples of an action that "may require additional review".
What this does not do
- Emails and phone numbers. Linked API reads what your LinkedIn account can see; it doesn't look up contact details.
- Intent data and call recording. Keep your data provider and your call recorder for those.
- CRM updates. ChatGPT doesn't write the call back to your CRM unless you also add a CRM plugin in ChatGPT, or build it with n8n or Make.
- Sequences. There are no scheduled follow-ups and no automatic stop when someone replies – each follow-up is a run you start and approve.
- Alerts. Nothing tells you when an attendee posts or changes jobs; ChatGPT reads LinkedIn when you ask.
- Private profiles. A private or limited profile stays limited – ChatGPT reads what your account can see.
Frequently Asked Questions (FAQ)
Use it where it has real context: briefing yourself before a call and drafting the follow-up after it. OpenAI's Sales plugin brings in your CRM, call recordings, Slack and Teams. Connected to your own LinkedIn account through Linked API, ChatGPT also reads the attendees' recent posts, who you know at their company and your conversation with each of them, and drafts LinkedIn replies you approve.
Preparing for and following up on real conversations, from real context. A generic prompt gets a generic opener; a brief built from the person's recent posts and your last exchange gets one they recognize. Keep the approval step: ChatGPT drafts, you decide what is sent.
Two ways. LinkedIn's own plugin in ChatGPT looks up professionals – per its listing, their title, company, location and follower count, with a link to the profile. Connecting your own LinkedIn account through Linked API adds what that account can see and do: your network at a company, your conversations, and replies and connection requests sent when you approve them.
The apps OpenAI's Sales plugin page lists are Slack, Teams, Zoom, Granola, Fireflies, Otter.ai, Salesforce and HubSpot; another 19 are counted but not named there. LinkedIn is not among the listed ones.
Per OpenAI's help, custom MCP server connections with full support are available on ChatGPT Business and Enterprise/Edu on the web, and on Business an admin or owner adds them. Pro gets read and fetch tools only; by OpenAI's rule and Linked API's read-only marks, it can look people and companies up and read messages already captured, but can't load missing conversation history or send. The setup is linked above.
Not in these runs: each prompt asks ChatGPT to show the drafts and wait, and nothing is sent until you approve. ChatGPT also has its own approval setting for plugin actions – with "Always ask", it asks before reading app information or making changes, sending a message included.
Yes – searching for the right people and writing first notes works the same way through your LinkedIn account; see AI sales prospecting on LinkedIn.
Yes. The same LinkedIn account connects to Claude, with Anthropic's Sales plugin covering the CRM side – see Claude for sales.
Build it in, or hand it to an agent
Get it out of the box: connect the MCP server as set up above, the agent-friendly CLI or a ready-made skill, and ask in plain language. Or build it into your own tools with the REST API, the Node and Python SDKs and the shell CLI. Either way it runs on your own LinkedIn account. Plans are flat per seat, from $49/mo billed annually – see pricing.
Facts verified November 9, 2026 – OpenAI's Sales plugin page, LinkedIn's ChatGPT plugin listing and OpenAI's help articles on custom MCP servers and action approvals checked on that date.