What Is a Good LinkedIn Ads CTR?
Search this and you get a consistent answer from a dozen benchmark pages:
| Figure commonly published | What it is usually described as |
|---|---|
| 0.44% – 0.65% | Average LinkedIn Ads CTR |
| ~0.52% | Median CTR for Sponsored Content |
| 0.50% – 0.65% | B2B SaaS Sponsored Content range |
| Below 0.45% | Usually read as "creative is not resonating" |
Those numbers are real, in the sense that people genuinely measured something and reported it. The problem is what they measured.
Almost none of those pages state which click field the figure is based on. And LinkedIn has two.
If a benchmark is built on clicks — which is the default, and which counts likes, comments, shares and profile clicks — then it is not a click-through rate at all. It is an engagement rate wearing a CTR label. Compare your honest landing-page number against it and you will conclude your account is failing when it is performing normally.
So what is actually good? On LP CTR — landing page clicks over impressions — a healthy B2B feed campaign generally sits in the 0.15%–0.35% range, with thought leader ads often lower because so much of the interaction is social rather than navigational. Single image ads to a tight, well-matched audience can run higher.
But the honest answer to "what is a good CTR" is a question: which clicks are you counting? Until that is settled, the number is not comparable to anything. The rest of this guide explains why, what it does to your other metrics, and how to fix your reporting.
The Field That Breaks Everything
LinkedIn's Marketing API returns a field called clicks. Here is what it counts:
- Clicks through to your landing page
- Likes
- Comments
- Shares
- Clicks on the poster's profile
- Clicks on the company page
- Clicks to expand "see more" on the post text
One of those is traffic. The other six are social engagement.
There is a separate field, landingPageClicks, which counts only the first one. It is the field you want, and it is not what Campaign Manager shows you by default. It is not what most dashboards pull. It is not what most agencies report.
This is not a secret or a bug. It is documented. It is just that almost nobody reads the field definitions, and the default is wrong in a direction that flatters everyone involved.
What It Does to Your Numbers
Take a campaign with £5,000 spend and 250,000 impressions. It generated 2,500 total clicks, of which 500 were actual landing page visits — a 5× ratio, which is unremarkable for a thought leader ad.
| Metric | How it usually gets reported | What is actually true | Distortion |
|---|---|---|---|
| CTR | 1.00% | 0.20% LP CTR | 5× overstated |
| CPC | £2.00 | £10.00 LP CPC | 5× understated |
| Cost per 1,000 visits | £2,000 | £10,000 | 5× understated |
| Implied LP conversion needed for £200 CPL | 20% | 4% | Wildly different job |
That last row is the one that matters. If you believe you are buying clicks at £2, you plan for a landing page converting at 20% and wonder why your CPL is four times target. If you know you are buying visits at £10, you plan properly — and you probably rebuild the landing page before you touch the campaign.
The error does not just misreport the past. It sends you to fix the wrong thing.
Why Thought Leader Ads Suffer Most
The gap scales with how social the format is.
A single image ad from a company page gets modest engagement, so total clicks and landing page clicks stay reasonably close — often 1.5–2×. A thought leader ad from a personal profile is designed to be engaged with. Likes, comments, profile clicks, "see more" expansions. That is the entire point of the format. So the ratio widens, routinely to 5× and sometimes to 10×.
Which produces the single most expensive mistake we see in account audits:
Compare formats on LP CPC and the picture is stable, because you are counting the same event in both cases: a human arriving on your site.
Why AI Makes This Worse, Not Better
Here is the part that is new in 2026.
There are now at least eight ways to connect a LinkedIn ad account to Claude or ChatGPT — MCP servers and connectors from data platforms, reporting tools and agencies. We compare them in the LinkedIn Ads MCP guide. Most of them pass LinkedIn's fields straight through. If the server hands the model a field labelled clicks, the model reports clicks. It has no way to know the label is doing something dishonest.
A spreadsheet containing a bad number looks like a spreadsheet containing a number. You might squint at it. An assistant containing a bad number produces this:
"Your thought leader ads are performing strongly, with a 1.0% CTR — roughly double the LinkedIn B2B average — at an efficient £2.00 CPC. I'd recommend shifting additional budget here."
Fluent. Structured. Reasonable-sounding. Completely wrong, and now carrying a recommendation attached to real money.
The failure mode is not that the model hallucinated. It did the arithmetic correctly on the data it was given. Bad data plus fluent writing is worse than bad data alone, because fluency suppresses the instinct to check.
The Other Three Metrics That Lie
Clicks is the worst offender, but it is not alone.
Lead gen form opens are not leads. The API returns oneClickLeadFormOpens and oneClickLeads. The first is someone tapping the ad and seeing a pre-filled form. The second is a submission. Report opens as leads and your CPL can look several times better than it is. Only submissions count.
Video views are not attention. LinkedIn counts a view at two seconds, which on autoplay means "scrolled past slowly". Track 25%, 50% and 75% completion instead. A campaign with 100,000 "views" and 400 quarter-completions has an audience of 400.
Conversation ad clicks are close to meaningless. The format lives in the messaging inbox. What matters is opens and open rate, then click-to-open. Judging a conversation ad on raw clicks is measuring the wrong stage of the interaction entirely.
The pattern in all four: LinkedIn's default metric measures activity on LinkedIn. Your business cares about activity that leaves LinkedIn. Those are not the same thing, and the platform has no particular reason to make the distinction obvious.
How to Fix Your Reporting
1. Rebuild your reports on landing page clicks. LP CTR and LP CPC become the primary numbers for every feed format. Keep total CTR if you like, labelled clearly as engagement, and never let it sit in the same column as a traffic metric.
2. Never label anything "CTR" when it is not landing page clicks over impressions. If it is total clicks over impressions, it is "engagement rate". Naming discipline is most of the fix.
3. Request fields explicitly if you use the API. The adAnalytics endpoint returns only impressions and clicks unless you name every other field in the fields parameter. Plenty of dashboards were built on two metrics because nobody realised the rest were opt-in.
4. Re-baseline before you panic. Your LP CTR will look terrible next to the CTR you have been reporting. It is not worse performance, it is the first honest measurement. Benchmark against LP CTR figures, not against a mixed number.
5. Ask your AI assistant for the right field by name. "Landing page clicks", not "clicks". Then ask which tool it called and what date range it used. A good session is auditable.
"But Engagement Has Value"
It does. This is the fair objection and it deserves a straight answer.
Engagement on a thought leader ad is genuinely valuable. It is a trust signal, it drives reach, and the people who like and comment are often the ones who later convert. Engagement data is also how you build the warm outbound lists that make LinkedIn worth running at all.
The argument is not that engagement is worthless. It is that engagement and traffic are two different things and must be counted separately. Blending them into one number destroys both. You cannot optimise engagement if it is hidden inside a click count, and you cannot forecast pipeline from traffic if that traffic is inflated fivefold by likes.
Measure both. Report both. Never add them together and call the total "clicks".