What KPIs should a LinkedIn ads agency actually report on?

Most LinkedIn ads reports open with impressions, click-through rate, and cost per lead. Those numbers are real, they are just answering a question B2B buyers did not ask: they measure whether the ad ran, not whether the people who decide deals noticed you.

Here is a reporting structure that survives a CFO conversation.

Tier 1: did we reach the right people?

Target-account reach. What share of your named account list saw the campaign at all. This is the single most important number in account-based LinkedIn advertising and the one most reports omit.

Audience composition of impressions. The percentage of impressions delivered to in-profile roles at target accounts. Organic distribution typically puts 10 to 15 percent of impressions in front of the intended audience. Well-targeted paid amplification should push that to 85 to 90 percent. If your paid number is not dramatically better than your organic number, targeting is broken.

Frequency per account. Reach without repetition does not build familiarity. Track average impressions per target account per month, not just per person.

Tier 2: did they engage?

Target-market engagements. Reactions, comments, shares, and clicks from people at target accounts, counted as people rather than events. A post with 25 interactions from buyers beats one with 2,000 from strangers.

Cost per target-market engagement. Total spend divided by identified engagements from target accounts. This replaces cost per click as your efficiency metric. In our programs, a $500 Thought Leader Ad budget on the right post has produced over 200 target-market engagements, which is a very different unit economics story than cost per click implies.

Accounts with multiple engagers. Two or more people from the same account engaging in a 30-day window is one of the earliest reliable buying signals on LinkedIn.

Repeat engagers. Individuals engaging with three or more posts. These are the people outbound should be contacting first.

Tier 3: did it influence revenue?

Pipeline influence rate. The percentage of CRM opportunities containing at least one contact with a recorded LinkedIn interaction. One Influent client reached 38 percent verified pipeline influence in the first year.

Engaged-account to opportunity rate. Of the accounts that engaged this quarter, how many entered pipeline within two quarters.

Sales-cycle and win-rate comparison. Deals with LinkedIn-engaged contacts versus deals without. This is usually where the strongest number in the whole report lives.

Self-reported attribution. Add "how did you hear about us" to your forms and record it on discovery calls. Imperfect, and still the closest thing to truth for content that influences buyers who never click.

Metrics to stop paying attention to

• Impressions on their own. Volume without composition tells you nothing.
• Click-through rate as a primary KPI. Thought leadership frequently succeeds without a click. High CTR often just means the creative promised something transactional.
• Cost per lead from lead-gen forms. Cheap leads from non-target accounts make the report look good and the pipeline look worse.
• Follower growth. A vanity metric unless the new followers are your buyers.
• Engagement rate. A ratio that rewards small audiences and punishes reaching new people.

What a good monthly report looks like

• Spend by campaign and by post.
• Target-account reach and coverage against the named list.
• Percentage of impressions delivered to target roles and accounts.
• Named people and companies from target accounts who engaged.
• Accounts with two or more engagers this month.
• Cost per target-market engagement, trended.
• Pipeline influence rate and its change.
• A plain-language read of what changed and what will change next month.

Reporting cadence that works

Weekly for spend pacing and delivery checks. Monthly for engagement composition and named-account movement. Quarterly for pipeline influence, win rate, and sales-cycle comparison. Judging pipeline monthly in a six-month sales cycle produces bad decisions in both directions.

One honest caveat

LinkedIn does not reveal everyone who saw a post. Many senior buyers read without ever reacting, and saves and some profile views stay anonymous. Buyer-level engagement data is directional evidence of visible attention, not a complete record. That is exactly why pipeline influence and self-reported attribution belong in the same report as engagement metrics.

Adjacent reading: the Thought Leader Ads playbook and how to measure category ownership.

Part of the series: this guide sits under the best LinkedIn marketing agencies in 2026, which compares the whole market by job.