LinkedIn Ads Benchmarks: What Good Looks Like in B2B

Most LinkedIn ads benchmarks are quoted as one number. A good click-through rate is 0.5%. A good cost per click is around ten dollars. Under that, you are fine. That single number is the problem. In the best public dataset we could find this year, median CTR ran from 2.68% on one ad format down to 0.02% on another. Same platform, same dataset, same year. A 134x spread.

So the useful question is not whether your CTR is good. It is which row you should be reading. This piece gives you the format-level numbers and the arithmetic that ties them together. It also gives you a check to run on any benchmark table somebody hands you. For what LinkedIn actually costs and how to budget a test, we wrote that up separately in LinkedIn Ads costs for B2B.

LinkedIn does not publish a benchmark table

Start with the awkward part. LinkedIn itself does not hand you a public set of performance benchmarks. We went looking on 24 August 2026 and did not find one. Their own Thought Leader Ads product sheet, the two-pager their sales teams send, carries no CTR, no CPC, no engagement rate and no sample size. Not a single performance figure.

That absence explains the mess downstream. Most benchmark tables you have read were assembled by somebody else, usually from their own book of accounts. No two books hold the same advertisers. One agency’s clients are enterprise software in the US. Another’s are recruiters in Poland. Both publish “LinkedIn benchmarks” and both are telling the truth about their own data. Neither is telling the truth about yours.

The one dataset this piece uses

We work from a single source here rather than average four of them. Averaging samples you cannot inspect produces a number that describes nobody. The one we picked is the ZenABM 2026 LinkedIn ABM Performance Benchmarks report, published 22 April 2026. We read it and checked the figures against the page on 24 August 2026.

Its stated basis: real performance data from 211 B2B companies, 161,256 LinkedIn ads, $5.5M in ad spend, and campaigns across 29 countries. The report says this is Campaign Manager performance joined to CRM pipeline, not survey answers. That matters. A survey asks marketers to remember their CTR, and marketers remember their best quarter.

Two honest limits before we use any of it. It is one vendor’s book of accounts, weighted toward account-based programmes, so it is not the industry. And the report itself says its coverage of Message Ads is thin. Read what follows as one large, well-described sample, not as the truth about LinkedIn.

Median click-through rate by ad format

Here is what the report found, format by format. These are medians rather than averages, so one enormous campaign cannot drag a row upwards on its own.

Thought Leader Ads sit at 2.68%. Event Ads at 0.55%. Document Ads at 0.43%, Single Image Ads at 0.42%, Carousel Ads at 0.32%, Video Ads at 0.24%. Text Ads, the small unit in the right rail, come in at 0.02%.

Data table

Ad format	Median CTR
Thought Leader Ads	2.68%
Event Ads	0.55%
Document Ads	0.43%
Single Image Ads	0.42%
Carousel Ads	0.32%
Video Ads	0.24%
Text Ads	0.02%

Look at the middle of that list. Document, Single Image and Carousel all sit between 0.32% and 0.43%. Three different production efforts, one flat line. Then Thought Leader Ads, which run a real person’s post rather than a company post, sit six times higher than any of them.

The gap between formats beats the gap between advertisers

This is the part that changes how you read your own account. Take two formats you would genuinely choose between, thought leader ads and single image ads. Their medians differ by 6.4x. Across the client accounts we run, careful work against careless work inside one format has never opened a gap that wide. Format is the bigger lever, and it is the one you set once.

So the format you picked in an afternoon can matter more than a quarter of tuning. Say you run single image ads at 0.45% CTR. You are above the median for your format. You are also at about a sixth of the thought leader median. Headline testing does not close a gap that size.

It also means the benchmark question most teams ask has no answer. “Is 0.4% a good CTR on LinkedIn?” Good compared to what. For a single image ad, that is average. A thought leader post at 0.4% would be a bad month. Text ads would call it twenty times the median. The number alone carries no information.

You are not paying for impressions, you are paying for clicks

Now put cost next to it. The same report gives typical CPMs by format: Single Image Ads at $59.15, Thought Leader Ads at $49.37, Carousel Ads at $45.28, Video Ads at $38.94. Text Ads sit around two dollars, which the report gives as an approximation, so we are leaving them out of the comparison.

Data table

Ad format	CPM
Single Image Ads	$59.15
Thought Leader Ads	$49.37
Carousel Ads	$45.28
Video Ads	$38.94

Notice how little movement there is. Across those four formats, the price of a thousand impressions varies by a factor of 1.5. Over the same four formats, median CTR varies by a factor of 11. Impressions are priced inside a narrow band. What changes, and changes a lot, is how many people click one.

That is why the CPC column looks the way it does. The report gives Thought Leader Ads at $2.29, Single Image at $13.23, Carousel at $13.30 and Video at $15.61. Video is the most expensive click in the list while being the cheapest impression in it. The impression price did not do that. The click rate did.

The check to run on any benchmark table

Two minutes and a calculator will tell you more about a benchmark table than the numbers in it. The three metrics are not independent. Cost per click is cost per thousand impressions divided by clicks per thousand impressions, and clicks per thousand is just CTR times ten. So:

CPC = CPM / (CTR x 10)

Run it on this report. Thought Leader Ads: $49.37 divided by 26.8 gives $1.84, against a reported $2.29. Single Image: $59.15 divided by 4.2 gives $14.08, against $13.23. Carousel: $45.28 divided by 3.2 gives $14.15, against $13.30. Video: $38.94 divided by 2.4 gives $16.23, against $15.61.

Data table

Ad format	Implied CPC	Reported CPC
Thought Leader Ads	$1.84	$2.29
Single Image Ads	$14.08	$13.23
Carousel Ads	$14.15	$13.30
Video Ads	$16.23	$15.61

Three of the four land within 6%. The thought leader row is 24% out. Close enough to say the rows describe one population. It is also as close as you should expect, for a reason worth knowing: medians do not compose. The median of a ratio is not the ratio of the medians, and this report labels its CPM figures typical rather than median. Perfect agreement would actually be the suspicious result.

What you are looking for is the table that fails badly. Say a table’s CTR and CPM rows imply a cost per click three times the one it prints. Something in there is not from one sample. Rows get stacked into a table without anyone checking that they agree. That happens more than you would like, and this is how you catch it in two minutes.

What good looks like for your account

Benchmarks are for sanity, not for targets. Three rules we work to at COSEOM®, and they are unglamorous.

Compare inside a format, always. Your single image campaigns get held against single image medians. Blend formats into one account CTR and that number moves whenever your format mix moves. You will spend a month explaining a change nobody made.

Then compare against your own history before anyone else’s. Your audience, your offer and your market set your floor long before your creative does. Twelve weeks of your own numbers is a better benchmark than a stranger’s median, because it holds those three things still.

Use the outside number for one job only: deciding whether a format is worth trying. A six-fold median gap is a reason to test thought leader ads next quarter. It is not a promise that yours will do 2.68%. Write that figure into a forecast and you will be explaining it in November.

Where a good benchmark still sends you wrong

One more caution, and the report makes it itself. A high click-through rate can go with worse pipeline, not better. The report flags that pattern where targeting is broad. Our own reading of why: widen an audience and you add people who are curious rather than buying. The CTR column goes up. The deals do not follow.

The same report notes that LinkedIn’s influence gets undercounted without account-level tracking, which pushes in the opposite direction. So one campaign can hand you a flattering engagement number and an unflattering pipeline number in the same week. Both can be wrong.

The way out is not a better benchmark. Pick the metric that matches what the campaign was for. Decide it before launch. Then let the benchmark tell you one thing only: whether your number is strange. If you want a second read on your account, that is what our LinkedIn advertising work is for.

FAQ: LinkedIn ads benchmarks

What is a good CTR on LinkedIn ads?

It depends entirely on the format. In the 2026 ZenABM dataset the median was 0.42% for single image ads, 0.43% for document ads and 2.68% for thought leader ads. A single “good CTR” figure for LinkedIn does not exist, because those rows are not describing the same product.

Why do published LinkedIn benchmarks disagree with each other?

Because each one is built from a different book of accounts. LinkedIn publishes no official benchmark table, so the figures you read come from someone else’s book: an agency’s clients, a platform’s users, or a survey of marketers. Two honest reports can differ by several multiples.

Does LinkedIn publish official ad benchmarks?

Not publicly, as far as we could establish on 24 August 2026. LinkedIn’s own Thought Leader Ads product sheet contains no performance figures at all. Campaign Manager will show you your own account’s numbers, but that is your data, not a public benchmark.

Why are thought leader ads so much cheaper per click?

Not because the impressions are cheap. In this dataset their CPM sat at $49.37, close to the middle of the pack. Their median CTR was 2.68%, six times single image ads, and a click rate that high divides the same impression cost across far more clicks.

How do I check whether a benchmark table is internally consistent?

Divide the CPM by the CTR times ten and see whether you get the printed CPC. Small gaps are expected, because medians do not compose exactly. In the dataset used here, three of the four rows landed within 6% and the fourth was 24% out. A factor of two or three is the signal that those rows are not describing one sample.

Should I compare my account to industry benchmarks or to my own history?

Your own history first, for anything you plan to act on. Your audience, offer and market move performance more than creative does, and only your own account holds those steady. Use outside benchmarks to decide whether a format is worth testing.

Is a higher click-through rate always better?

No. The ZenABM report flags that a high CTR can come with lower pipeline quality where targeting is broad. Read the click rate next to what happened after the click, or you will optimise toward an audience that enjoys your ads and never buys.

Which LinkedIn ad format had the lowest median CTR?

Text ads, at 0.02%. Video came second lowest at 0.24% while carrying the highest cost per click in the dataset at $15.61. Low click rates are not automatically a reason to drop a format, but they should change what you expect it to do.

How often should benchmarks like these be refreshed?

Treat any table older than about a year as a rough direction rather than a number. Ad formats change, auction pressure changes, and this dataset itself was published in April 2026. The relationships between the metrics hold up far longer than the values do.