The LinkedIn Metrics That Actually Predict Leads

Why impressions and likes fail as a lead signal

Impressions recount how many feeds your post appeared in. Likes tell you how many strangers scrolled, stopped, and gave a one tap gesture. Neither number knows whether your buyer saw the post. When a founder posts an in-depth teardown of a real migration and gets 9 likes, and the next day posts a broad opinion on hiring and gets 180 likes, the second post "wins" the dashboard and loses the pipeline: the 180 were other founders, students, and job seekers, and none of them will ever buy a $4,000 per month retainer. Scores like these feel productive and train you away from the narrow content your buyers need.

There is a further distortion: engagement from fellow content creators (other ghostwriters, marketers, engagement pod members) stacks your numbers while teaching you nothing about the buyer market. The cleanup: every metric you act on must be filtered down to "did it come from a target company or a target title?" Anything else is noise you can ignore without a second thought.

The five metrics that actually matter

These are the numbers we report to founder clients every week. They are smaller numbers, and they move the needle.

1. Qualified profile visits. Weekly count of visits to your profile from people who match your ICP (the viewer's title or company is visible in your analytics). This is the metric that says "the content is putting you in front of exactly the right strangers," and it is the best leading indicator of inbound asks we know.

2. Comments from target titles. Not total comments: comments from VPs, directors, founders, and owners in your niche. Ten targeted comments beat 400 unknown comments in every week we have measured it.

3. Inbound asks. Counted by type: question about how you work, ask for a resource, ask for a call. Three inbound asks a week is a thriving founder account long before the reach numbers convince anyone.

4. DM replies from warm leads. How many of your sent messages earned a substantive reply this week. This tells you whether your outreach list is fresh and whether your feed made you recognizable before the DM arrived, because warm reply rate rises with months of consistent posting.

5. Calls booked, by source. The lagging metric every other one exists to grow. Tag every call by origin (LinkedIn post, DM, referral, other) and count monthly. If calls are not rising after a quarter of consistent execution, do not blame the metrics; change the content mix, because the diagnostic always shows the missing link among numbers 1 through 4.

A weekly 20 minute review routine

Metrics worth reading weekly, in the same order, at the same time (Friday morning works well for us):

1. Open your analytics and count qualified profile visits and target title comments. Log both in one sheet.

2. List every inbound ask from the week, with its source. One DM that mentions a post by name is a grade A signal; write down which post it named so that content type earns a repeat.

3. Note the single best performing post by the two relevant metrics, and the worst, and write one sentence explaining each in terms of audience rather than luck. (Formats you can see: "explained a concrete decision" beats "shared an industry mood.")

4. Pick one change for next week based only on evidence: adjust the posting time, sharpen the hook toward the buyer's task, double down on the topic that pulled the target-title comment. One change per week; three measures churn that makes everything unmeasurable.

This routine takes 20 minutes a week and replaces most of what people think a "quarterly analytics deep dive" does. Founder accounts do not fail from lack of analytics tools; they fail from reading the numbers that did not matter.