Account intelligence from LinkedIn engagement: roll engaged people up into account summaries

Account intelligence from LinkedIn engagement means grouping the individual people who engage with relevant posts by the company they work for, so you can see which accounts have several qualified people paying attention, how well they fit, and how recent the activity is. In traxy, qualified leads carry their company, ICP match and signal. You can roll them up by account through the public API or MCP account reports, which list account-level summaries from visible active leads and sort by latest signal, lead count, average ICP match or account name. In the app, you filter Leads by company and save the view. Use the roll-up to choose accounts, then contact people using their own signal.

Last checked: November 9, 2026, against traxy's MCP tools, REST API reference and lead docs.

Is there a tool that alerts you when several people from the same target account engage on LinkedIn?

Not as one built-in alert: traxy has no built-in alert on the number of people per account. traxy alerts you in Slack for each qualified lead, shows the company and the post or comment behind every lead, and groups leads by company under Leads > Companies, so you can see which accounts have several engaged people. A rule like "3 distinct people from one account in 14 days" is a count you set up yourself from traxy's lead events, sent to a webhook, n8n or Zapier.

traxy runs your entire LinkedIn GTM end to end, from finding high intent leads from social intent to enriching and researching each one, to running relevant, personalized LinkedIn and email outreach to them on autopilot once you approve the campaign. This section covers both halves: spotting the accounts where several people engage, then reaching those people.

The workflow in traxy

1. Track what your target accounts read. Add competitors, industry voices and your own team under Targeting > People, company pages under Targeting > Companies, and your signals under Targeting > Signals (tracking).

2. Let traxy qualify the engagers. Only people with a 50%+ ICP match become leads, and each one keeps the post or comment behind it.

3. Get a Slack alert for each lead. Connect Slack under Integrations, set the default channel and add a route from a saved lead view (Slack, routes). Slack can also post repeat-engagement events, which are new activity from a lead who already qualified.

4. See which accounts have several engaged people. Open Leads > Companies. Each row shows the matching people count, best ICP match, latest signal, industry and size, and expanding a row shows its people (companies). In the people table you can also show the company domain column and sort by it, or filter by company and save a shared view per target account list.

5. Or ask your AI client. The account reports described below sort accounts by lead count.

Set up a "3 people in 14 days" account alert yourself

traxy does not count distinct people per account or fire an alert at a threshold, so the count lives in your own workflow:

1. Connect a Webhook, n8n or Zapier destination under Integrations and add a route from your lead view with Auto delivery. Automatic routes apply to leads entering the view from then on, so send existing leads manually with Send to if you want them counted (routes).

2. Map the company field. Lead events are signed and carry lead, company, contact and engagement fields, and first-delivery and repeat-engagement events are separate (webhook). Use the payload sample in the app to find the company domain.

3. Store each event with the company domain, the person and the date, then count distinct people per domain over the last 14 days.

4. When the count reaches your number, post a message to Slack or create a task in your CRM. Delivery can retry temporary failures, so count people, not events.

The n8n LinkedIn lead workflow shows how to branch on traxy's event types in n8n. Pick the number and the window yourself: three people in 14 days is a starting point to test, not a benchmark, and the count only includes people who qualified as leads.

A real example of the per-lead half: Ben Buaron (traxy, Windmill Growth) set traxy up for 35 of his agency's clients to flag ICP-fit engagers and route them to Slack channels shared with each client's sales team, and says churn dropped (his post).

Then reach the people at that account

The alert is only the first half. Put the account's leads into a campaign through a shared view, and traxy researches each lead, groups leads by signal and prepares a personalized LinkedIn and email sequence for each group, sent from your own connected sender accounts. You test it on sample leads with nothing sent and review who joins and who is skipped at launch. Then the approved campaign runs on autopilot, or you approve every message in a review queue. Replies, unsubscribes and hard bounces stop the sequence.

You can run the same campaign from Claude, ChatGPT or another AI client through the traxy MCP, which waits for your yes in chat before anything sends. Message each person about their own engagement, not their colleagues' (see the worked example below). More on the outreach half in LinkedIn outbound with traxy.

Why roll engagement up to the account?

Person-level signals answer "who is paying attention." Account-level roll-ups answer three questions people-level views hide:

• Is this one curious person, or several people at the same company? Three qualified people at one company engaging with posts about the same problem is different from one.

• Which roles are involved? A head of sales and a RevOps lead at the same account suggest a different conversation than two SDRs.

• Is the activity recent? A cluster from last week matters more than one from last quarter.

This is useful for account-based teams and for anyone selling to a buying group rather than a single person. Our LinkedIn ABM guide covers the wider approach.

Account-level intent vs account summaries from engagement

These sound similar and work differently.

More on the distinction in account-level vs contact-level intent data. LinkedIn's own Marketing API also offers account-level engagement reporting for advertisers; that is aggregate ads data, separate from what this guide covers.

What an account summary should contain

Keep it short enough for a rep to read in a minute:

1. Account name

2. Number of qualified people engaging

3. Average ICP match across them

4. Latest signal date

5. Top people: name, role, signal label, and what they engaged with

6. Next step: who to contact first, and with what reference

Leave out anything you can't trace to a recorded signal. A summary that says "high interest" without the people behind it is a guess.

How to build account summaries in traxy

Through the API or MCP

traxy's public catalog includes two account tools for AI clients connected over MCP (MCP tools):

• traxy_list_account_reports: list account reports

• traxy_get_account_report: read one account report

The REST equivalents are GET /customer/account-reports, described as account-level summaries derived from visible active leads, and GET /customer/account-reports/{accountId}, one account summary with top visible leads. Per the API reference, the list accepts filters for company name, minimum and maximum ICP match, a date range and a search query. It sorts by latestSignalAt (the default), leadCount, averageIcpMatchPercent or accountName. Both need the leads:read scope (authentication and scopes).

An example request from Claude connected to traxy:

"List my traxy account reports from the last 30 days with ICP match of at least 70, sorted by lead count. For the top five, show the top leads and their signals."

Pass the workspace explicitly on every call (MCP workspaces). See find and review LinkedIn leads from Claude for setup.

In the app

1. Open Leads and filter by company, signal and workflow state (filter and save audiences).

2. Scan for companies that appear more than once.

3. Save a shared view per target account list so the team works from the same audience.

4. Open each lead's Signals tab to see what the person engaged with.

To keep watching a specific company, use Targeting > Companies > Track company (tracking). Company tracking is separate from the people you track individually.

Feed the roll-up with the right sources

An account summary is only as good as the engagement behind it. Useful sources:

• Competitors, tagged Competitor, so you see target-account people engaging with alternatives

• Industry voices your buyers follow, tagged Influencer

• Your own team's posts, tagged Teammate

• Specific posts you submit by URL for a one-off look at who engaged

Ongoing monitoring for a tracked profile follows the five most recent posts published after you add it. For older posts, submit them as tracked posts. The who to track on LinkedIn guide goes deeper.

Worked example

A sales consultancy sells outbound programs to software companies.

Its account report, sorted by lead count, shows one company with four qualified leads in the last 30 days: a VP of Sales, two sales managers and a RevOps manager. Three of them commented on posts about SDR ramp time, and their average ICP match is high.

The team decides to contact the VP of Sales first, referencing the VP's own comment. The RevOps manager gets a separate message about reporting, referencing their engagement. The team does not tell either person that colleagues engaged, because that reads as surveillance and the person didn't share it.

This example shows the workflow. It does not report a result.

Using account summaries in outreach

• Choose accounts with the roll-up. Write messages from the individual signal. Each person should only hear about what they did.

• Sequence by role. Start with the person closest to the problem or the budget, and avoid messaging four people at once.

• Check do not contact first. Suppression applies per person (do not contact).

• Review before launch. In a campaign, the launch review shows who joins and which leads another unfinished campaign holds (review and launch).

Send account context to your CRM

If reps work from a CRM, route a saved view through Zapier or a webhook and write the signal into the record (routes). Then group by company in the CRM. For HubSpot options, see best LinkedIn tools for HubSpot.

Limits

• Account reports are built from visible active leads. People who engaged but did not qualify will not appear, and leads you moved out of active work may not either.

• Engagement shows relevance. It does not prove a buying decision or budget.

• A missing signal at an account means you don't know. It isn't evidence of no interest.

• Several posts can include the same person, so post-level counts don't equal unique people.

FAQ

Is an account report the same as an ABM platform's account score?

No. It summarizes named, qualified leads and their signals. It does not score anonymous company research activity.

Can I see account reports in the traxy app?

The account report tools are documented in the public API and MCP catalog. In the app, use Leads filters and saved views to group people by company.

Does tracking a company add its employees as leads?

Company tracking is documented as separate from tracking individual people. Leads still have to qualify against your lead criteria.

Does traxy alert me when several people from one account engage?

No built-in alert counts people per account. traxy alerts you in Slack for each qualified lead, groups leads by company under Leads > Companies, and sends lead events with company fields to a webhook, n8n or Zapier, where you can count distinct people per account and alert on your own threshold.

Can I see which accounts have several engaged people in the app?

Yes. Leads > Companies groups leads by company with a matching people count, best ICP match, latest signal, industry and size.

Which sort should I use?

Start with latestSignalAt for timing, then check leadCount for breadth and averageIcpMatchPercent for fit.

Next step

Connect an AI client to traxy, pull your account reports sorted by lead count, and pick three accounts where more than one qualified person engaged this month. Then plan one message per person, each built on that person's own signal.