How to track LinkedIn automation performance in a dashboard

Move LinkedIn outreach data into a custom dashboard with an API so you can track pipeline stages, reply intent and meetings without manual exports.

Get started

Short answer

To track LinkedIn automation performance in a dashboard, pull data from your automation tool into a spreadsheet or database through its API. Then map each lead to a stage and a reply intent, and chart the stages that lead to meetings. The dashboard should show pipeline movement. Message volume alone will not tell you where to change the script.

Decide what metrics actually matter

An automation tool will show you how many invites went out, how many messages were delivered and how many profiles were viewed. Those numbers are easy to collect and hard to use. If you send more messages and the same number of meetings come back, the extra volume did not help you.

Pick metrics that change a decision. A positive reply tells you the targeting and the opening line worked. A booked meeting tells you the conversation moved. A lead that replied but never booked tells you where the follow-up needs work. Track those stages separately so you can see where people stop.

Add a small set of health metrics: pending invites, replies waiting for an answer and leads with no next step. Health metrics tell you what needs attention today. Pipeline metrics tell you whether the campaign is worth continuing.

Activity metrics that feel productive

  • Total invites sent
  • Total messages delivered
  • Profile views
  • Connection request acceptance rate

Pipeline metrics that change decisions

  • Positive replies
  • Meetings booked
  • Leads that replied and did not book
  • Reply intent

Set up your data destination

You need a place where you own the rows and can change the logic. A spreadsheet is enough for one person. A database is better when several people update the same data or when you want to keep history.

Make one row per lead. Use the LinkedIn profile URL as the key. Names change, headlines change and people apply to the same campaign twice. The URL is the stable part.

Decide the stages before you import anything. If one person calls a lead 'replied' and another calls it 'engaged', the dashboard will split the same state into two columns. Write the stage names down and use them everywhere.

  1. 1

    Create the lead table

    Add columns for profile URL, name, company, campaign, current stage, reply intent, last action date and next step. Keep the columns narrow at first. You can add fields after you know what you will use.

  2. 2

    Use the profile URL as the primary key

    Set the URL column to reject duplicates. When a new export arrives, update the existing row instead of adding a second copy of the same person.

  3. 3

    Write the stage definitions

    For example, Invited means a connection request is pending. Connected means the request was accepted. Replied means a message came back. Interested means the reply asked about a call, a price or a next step. Booked means a meeting is on the calendar.

  4. 4

    Add a source column

    Record whether the lead came from a search, a competitor post or a CSV import. This lets you compare lead groups later without guessing.

Connect your automation tool via API

An API lets a script ask your automation tool for data on a schedule. You do not need a full engineering team. A short script that runs once a day and writes to a spreadsheet is enough to start.

Read the tool's documentation and find the endpoint that lists leads and the endpoint that lists replies. Pull the fields you need, not the whole record. Store the raw response as well as the mapped row so you can fix mistakes without asking the tool for the data again.

Put the key in an environment variable or a secrets manager. Do not paste it into a shared document or a browser script. If the key leaks, rotate it before you do anything else.

Keep the API key out of the dashboard. A read-only key for reporting is safer than the key your sending tool uses to change campaigns.

Paste this job

GET /leads?updated_since=2025-01-01
Authorization: Bearer YOUR_API_KEY

# Map each lead to one stage and one intent.
for lead in response['leads']:
    stage = 'Replied' if lead['last_reply_at'] else 'Connected'
    intent = 'Interested' if lead['reply_intent'] == 'positive' else 'Other'
    upsert_row(lead['profile_url'], stage, intent, lead['campaign_id'])

Build the visual dashboard

Build the funnel first. Plot total leads, connections, replies, interested replies and meetings. A drop between two stages tells you where to look. A drop after connection means the first message is weak or the offer is wrong for that group. A drop after interest means the follow-up or the scheduling step is losing people.

Add a time view next. Group by week and show new leads, replies and meetings. This helps you see whether a change to the script made a difference or whether a slow week was just a slow week.

Segment by campaign and lead group. A campaign that works for one group can fail for another. If you mix them in one chart, the average hides the useful part.

Leads

Funnel stage
Count of new rows by week
What to plot
The search or import is not bringing in enough people

Connections

Funnel stage
Share of invites accepted
What to plot
The profile or the invite note is not landing

Replies

Funnel stage
Count of leads with any reply
What to plot
The opening message is not giving people a reason to answer

Interested replies

Funnel stage
Count of replies with buying intent
What to plot
The offer or the qualification is off

Meetings

Funnel stage
Count of booked calls
What to plot
The follow-up or scheduling link is losing people

Keep the dashboard clean and trustworthy

Clean data is the difference between a dashboard you trust and a chart you argue with. Run a few checks before you look at the funnel.

  • Every row has a profile URL and a stage.
  • Every reply has an intent and a date.
  • No lead has two open stages at the same time.
  • Every booked meeting has a lead row behind it.
  • Every campaign has a send window and an owner.

A dashboard that is a day behind is still useful for trends. It is not useful for a reply that needs an answer now.

Use the dashboard to make decisions

Review the dashboard once a week and make one change at a time. If replies are coming in but meetings are not, read the last few threads. You may be answering questions without asking for a call. If meetings are booked but they are poor fits, check the lead group and the qualification in the opening message.

Use the dashboard to kill campaigns, not to defend them. A campaign with activity and no meetings is a campaign to pause. A lead group with replies and no bookings is a group to rewrite. The point of the dashboard is to help you stop doing work that does not move anyone.

Keep a short note next to each change. Write the date, the campaign and what you changed. When the next review comes, you can see whether the change helped without relying on memory.

Doing this with Kinlead

Kinlead has a REST Agent API and an MCP server. Chat apps connect with OAuth and scripts use API keys. You can use either to find leads, check replies and draft answers from Claude, ChatGPT, Cursor or another MCP client.

Kinlead does not send outbound webhooks, so a reporting script polls the API on a schedule. Pull the leads and replies you need, map them to your stages and write them to your own table. The Agent API and MCP page covers the connection options.

Frequently asked questions

Related

Get started with Kinlead

Prefer markdown? How to track LinkedIn automation performance in a dashboard.md