Most agencies are not dishonest. Most of these problems come from defaults nobody changed and metrics nobody questioned. But you are paying for the outcome, not the intent — and every one of these is checkable.
1. They report "clicks" rather than landing page clicks
This is the big one. LinkedIn's clicks field counts likes, comments, shares, profile clicks and company page clicks alongside actual visits to your site. On thought leader ads the gap between total clicks and landing page clicks routinely runs 3–10×.
If your report shows a CTR around 1% and a CPC that looks pleasingly low, you are almost certainly looking at total clicks. Measured honestly, your LP CTR is probably nearer 0.2% and your true cost per visit five times what you were told.
How to check: ask "is this landing page clicks or total clicks?" If they do not immediately know the difference, every report you have ever received has been inflated. Full explanation in what a good LinkedIn Ads CTR actually is.
2. Audience expansion is switched on
Audience expansion lets LinkedIn show your ads to people "similar to" your target audience. It is on by default. It is a monetisation feature, not a targeting feature — it exists to spend your budget on inventory you did not ask for.
On a tightly defined B2B audience it systematically dilutes the thing you were paying for.
How to check: Campaign Manager → any campaign → Audience → look for "Enable Audience Expansion". If it is ticked on a targeted B2B campaign, ask why.
3. The LinkedIn Audience Network is eating your budget
LAN places your ads on third-party apps and sites off LinkedIn. It is also on by default. Impressions are cheaper, which makes topline CPM look better, and the traffic quality is generally far worse — including a meaningful share on mobile games and low-quality apps.
For most B2B advertisers it is the largest single source of invisible waste, precisely because it improves the metrics people report on.
How to check: break your delivery down by placement. If a significant share of impressions are off-LinkedIn and nobody has discussed it with you, that is a conversation worth having.
4. There is no mid-funnel
Look at your campaign list. If it contains cold awareness campaigns and "book a demo" campaigns and nothing in between, you have no funnel — you have two disconnected campaigns.
The missing layer is 180-day retargeting segmented by engagement depth: video viewers at 25/50/75%, ad engagers, website visitors, people who opened a lead form and abandoned it. That is where warm audiences are built, and it is where most B2B accounts have nothing at all.
How to check: ask to see a retargeting campaign segmented by engagement depth. One catch-all "all warm audiences" campaign does not count.
5. Frequency is climbing while landing page CTR falls
Classic creative fatigue, and it is especially common on UK accounts where audiences are small enough to saturate in weeks.
When the same people see the same creative repeatedly, frequency rises and performance decays. Left alone, you pay increasing amounts to annoy a shrinking pool of people.
How to check: compare frequency and LP CTR over the last 30 days against the 30 before. If frequency is up and LP CTR is down, creative should have been refreshed already.
6. Lead gen form opens are being reported as leads
LinkedIn's API distinguishes oneClickLeadFormOpens — someone tapped the ad and saw a pre-filled form — from oneClickLeads, an actual submission. Reporting opens as leads makes cost per lead look dramatically better than it is.
How to check: compare the lead count in your report against the leads actually in your CRM. If the report says 200 and your CRM has 60, you have found it.
7. Spend is reaching job titles outside your ICP
Targeting is set up once and then drifts. Titles broaden, seniority filters turn out looser than intended, expansion widens everything. Six months later a real share of budget is reaching people you do not sell to.
It stays invisible because demographic data sits on a different screen from spend data and almost nobody reconciles the two against the target list.
How to check: ask for a spend breakdown by job title and seniority for the last 90 days, mapped against your ICP. A good agency has this already. On UK accounts specifically, watch for US seniority conventions being applied to UK titles — "Head of" is senior here and mid-level in an American org chart.
8. Every report leads with impressions
Impressions and reach at the top of a report usually mean the numbers further down are not good. Nobody leads with impressions when they have pipeline to show.
How to check: look at the order of your last three reports. If the first section is delivery volume rather than commercial outcomes, ask for the order to be reversed and see what happens.
9. Nobody can connect spend to pipeline
The final and most expensive one. If the reporting stops at cost per lead, you are measuring the top of your funnel and calling it results.
LinkedIn rarely gets last-click credit — B2B buyers take 15+ touchpoints over three to six months, and the final click is usually a branded search. An agency measuring only last-click leads will consistently under-report LinkedIn's actual contribution, and will optimise toward cheap leads that never close.
How to check: ask which closed-won deals had LinkedIn ad exposure before the opportunity was created. If nobody can answer, nobody knows whether any of this is working.
What to Do About It
Three or fewer of these is normal. Most accounts have a couple.
Five or more suggests the account is being maintained rather than managed. That does not automatically mean changing agency — it usually means having one uncomfortable conversation with specific evidence, and a decent agency will respond well to it.
You can check most of this yourself. The full structured version is the eight-layer audit, and if you would rather have an assistant do it in a minute, connect your account to the LinkedIn Ads MCP and ask it to run one.