Before you touch anything: the three questions
Most "LinkedIn doesn't work" verdicts get handed down in week three by someone staring at a lead count. Answer these three first, honestly. If any of them is a no, that's your answer and the rest of this page is a distraction.
Audience Network and Audience Expansion are both on when you open the account. Of the last 100 accounts we've reviewed, more than 70 had the Audience Network eating over half the budget. One of them got through £96,412 with 62% of that spent nowhere near LinkedIn, and came away with three demos and nothing closed. Check this before you look at anything else.
£10 a day is the platform minimum, which tells you nothing about what works. £500 a month usually buys 3 to 7 leads, so treat it as a test rather than a plan. £1,500 to £3,000 a month is where UK B2B results start being realistic. And if your average deal is under about £5,000, LinkedIn probably isn't the channel for you at all.
Give it 4 to 6 weeks and around 15,000 impressions before the algorithm has anything to work with. Don't judge a single ad on fewer than 2,000 impressions. Month 1 is rent, not results. Nobody should be calling it at day 30.
"The dashboard was green. The pipeline was empty. LinkedIn did exactly what it was told. The problem is nobody told it to find buyers."
Find your symptom
Pick the row that matches what Campaign Manager is showing you this morning. Not the thing you've been quietly dreading.
It is not spending, or it is spending on the wrong people
Delivery goes wrong two ways. Either nothing is serving, or loads is serving and almost none of it lands near a buyer. The second one is far more common and much more expensive, because the dashboard looks perfectly healthy while it happens.
- Look at matched members, not uploaded rows. LinkedIn wants 300 matched before it will serve you. A company list with LinkedIn URLs matches at 70 to 85%. Company names on their own, 40 to 60%. Contact lists on work emails, 20 to 40%. Upload 1,000+ companies if you want to be sure of clearing the floor.
- Give it 24 to 48 hours. Processing takes that long. Only re-upload if it's still stuck afterwards.
- Check the audience hasn't lapsed. Matched audiences expire after 90 days of no activity. Refresh at 60 and stop worrying about it.
- Check the bid. Manual CPC at about 80% of LinkedIn's suggestion is the right place to start, and it often ends up paying 30 to 50% less than suggested. Go much lower and you simply won't enter the auction.
- Kill the LinkedIn Audience Network. It puts your ads on third-party apps, mobile games and news sites. We've audited accounts where 40% of the budget went out that door with nothing to show for it, and one where it swallowed 62% of £96,412.
- Kill Audience Expansion. On by default, quietly serving people who vaguely resemble your target. It wastes more money than anything else on this page.
- Add the exclusions nobody bothers with. Students, entry-level, freelancers, industries you don't sell to, countries you can't ship to.
- Get suspicious when the CPC looks great. £1.50 clicks that never turn into a conversation cost you more than £9 clicks that produce a demo every fifty. In that £96k account, fewer than 400 of 8,742 clicks were the target audience. Reported CPC £11. Real cost per relevant click, north of £240.
"Ten clicks from the right people beat a thousand from the wrong ones. Every time."
Impressions are fine, nobody clicks
Before anything else: don't read this too early. An ad needs 2,000 impressions before its CTR means a thing, and you shouldn't pause one until it's had 2,000 with no clicks. Below that you're reading noise and calling it insight.
After that, it's blunt. CTR under 0.3% is a creative problem, not a bidding problem. Cold UK B2B normally runs 0.4 to 0.8% on single image and document ads, 0.6 to 1.2% on Thought Leader Ads. The UK and EMEA average sits around 0.60%.
Safe creative costs you 3 to 5x more per result, and it does it in a loop most advertisers never spot. Nobody stops for the ad. CTR drops. Delivery gets worse. CPC and CPL climb. You get twitchy and sign off something even safer next time. Round it goes.
- Open with the painful bit, not with what you sell. Nobody clicks a brochure.
- Be specific enough to put people off. An ad that could be sent to anyone lands with no one.
- Make the offer stupidly obvious. One thing to do, clear inside two seconds.
- Hit two of curiosity, emotion and clarity. That's the bar. One isn't enough.
- Try a Thought Leader Ad. The feed treats a face differently to a logo, and the CTR band shows it.
Then refresh on a schedule rather than in a flap. Every 4 to 6 weeks, or earlier if frequency creeps past 4 to 5 or CTR slides for a week straight.
Not sure whether it is the creative or the audience?
Book a 30 minute call and talk us through the account. We will tell you where we would look first.
Clicks are fine, conversions are not
Check tracking before you diagnose a single thing. Spent £500 and recorded zero conversions? It's almost never the offer. It's that nothing is being counted. We've audited accounts spending ten grand a month with no Insight Tag on the site whatsoever.
With tracking sound, the maths is easy. Cold LinkedIn traffic should convert somewhere between 2 and 5% from click to form fill, and a well-built page will do 5 to 15%. Under 1% and something's broken, and nine times out of ten it's the page rather than the ad.
- Insight Tag and conversion tracking. Installed, firing, wired to the right conversion action. Prove it before you touch anything else.
- The objective. Brand Awareness, Website Visits, Engagement and Video Views will never give you leads, however good the page is. You want Lead Generation, Website Conversions or Job Applicants.
- The five-second test. Show the page to someone who's never heard of you. In five seconds they should know what you do, whether it's for them, what it costs to carry on as they are, and what to do next. Most B2B pages fall over on the first one.
- It's talking to everybody. "Leading solutions for your business" converts nobody. "We cut IT downtime for manufacturers by 75% in 90 days" converts the right somebody.
- The CTA doesn't match the intent. High intent gets "Book a consultation" or "Get a quote". Middling intent gets a case study. Low intent gets a quiz or a subscribe. Put all four on one page and you've offered none.
- Message match. Whatever the ad promised, the page headline has to say it back. Make someone translate and they'll leave.
"Your website is talking to everyone, so it's converting no one."
Worth knowing before you rip the page up: 95 to 97% of B2B visitors leave without doing anything on their first visit. That's not a broken page, that's B2B. Once you've banked 500+ visitors, £10 to £15 a day on retargeting is usually the best-value spend available to you on the platform.
Leads arrive, sales say they are junk
Volume isn't your problem, so leave the budget alone. Junk comes from three places and you can fix all three without spending a penny less.
- Lead Gen Forms doing their job too well. Pre-filled forms get contact rates 20 to 30% below website forms. A form that takes 3 seconds to fill in produces a lead that takes 3 weeks to chase. A form that takes 30 seconds produces someone who answers the phone. Move that campaign to a landing page and take fewer, better leads.
- Audience Expansion, still on. This is why your named-account campaign keeps generating enquiries from firms nobody has ever heard of.
- Targeting titles instead of accounts. £500 a month at 200 named companies with two or three titles layered on will beat £5,000 sprayed across 500,000 people. Build the company list from your CRM, from a quarter of an hour of manual LinkedIn digging, or straight in Campaign Manager.
While you're in there, look at audience size. Under 50,000 is where precision lives. 50,000 to 150,000 is the CPC sweet spot at roughly £8 to £12. Anything over two million gives you £2 to £4 clicks and lead quality to match.
It worked, then it stopped
A campaign that fades is a different beast to one that never got going. Go through these before you write the channel off.
- Frequency above 5. The audience is too small for the money you're putting through it and people are sick of the ad. New creative, wider list. Don't kill the campaign.
- CTR sliding for a week or so. Ordinary fatigue. Refresh every 4 to 6 weeks as routine and it rarely gets this far.
- The calendar. July and August are the cheapest months on the platform. Q4 is the dearest and the most crowded. A December dip usually isn't your campaign's fault.
- Someone changed the site. New template, slower page, a form that's quietly broken. Compare landing page conversion rate over the same window before you blame the ads.
- You've run out of list. Tight named lists saturate faster than people expect. Add the adjacent segment or push the weight into retargeting rather than hammering the same 300 people again.
The patience problem: month 1 is rent, not results
A fair few of the accounts we're asked to rescue were never broken. Somebody switched them off on day 25. Three things happen under the bonnet in that first month and none of them show up on the dashboard: LinkedIn exploring the audience, creative pacing over the first fortnight, and bidding settling down around week 3.
- Day 8: £180 spent, 12,000 impressions, 41 clicks, 0 leads
- CPC bouncing around while bidding settles
- No retargeting worth running yet; you need 500+ visitors
- A 2 to 6 week gap between first click and first demo
- Tracking dead: no Insight Tag, no conversion recorded
- Audience under 2,000 or over 2 million
- Frequency above 5 in the first fortnight
- 200+ clicks and zero form submissions
- All ads saying essentially the same thing
Anything on the right, deal with it this week. Anything on the left, leave it. Give the channel 90 days before you pass judgement, and check the account weekly instead of every morning with your coffee.
"Month 1 is rent, not results. Week 1 is a setup signal, week 2 is the first read, week 3 is the pattern, week 4 is the first real decision."
What is actually normal (UK, 2026)
Half the "it's not working" diagnoses we hear disappear the moment someone checks the benchmarks. Here are the UK B2B ranges. Compare yourself before you conclude anything, and keep in mind that average isn't good. Hitting average means the quartile above you is beating you.
Not one of those numbers decides anything on its own. CPQP, cost per qualified pipeline, is the figure that belongs in a board meeting. An ugly CPL sat next to a healthy CPQP is a campaign you leave well alone. Do the deal maths before you panic: average deal value times close rate has to beat CPL times the leads you need. A £10,000 deal closing at 10% gives you about £1,000 per ten leads, so £100 CPL breaks even and £60 is a decent business.
Diagnose in order. Change one thing. Give it 2,000 impressions.
Delivery, then creative, then landing page, then lead quality, then fatigue. Nearly every account we take over that "doesn't work" is broken at one of those five points, usually the first or the third, and has been made considerably worse by somebody trying to fix all five in an afternoon.
LinkedIn's recommendations exist to fill LinkedIn's bank account. Somebody has to be the adult in the room, and it's you.
Tools to run the diagnosis yourself
Check whether your budget can reach your list often enough to learn anything.
Rebuild targeting around named companies instead of job title soup.
Fresh angles for when CTR says the creative is the problem.
Honest comparison of fixing this in-house versus with a specialist.
Common questions
If you would rather someone else diagnosed it
- A full account audit against the five failure points on this page, with the order we would fix them in.
- Settings sweep: Audience Network, Audience Expansion, exclusions, bidding and objective.
- Audience rebuild around a named account list, with match rates checked before anything goes live.
- Three new creative angles built from your own material, not stock imagery.
- Landing page and tracking review, because most 'ads problems' turn out to be landing problems.
Starter £975/mo, Growth £1,725/mo, Scale £4,025+/mo. Most accounts we take over need the settings, the target list and the landing page fixed before any spend increase makes sense.
Your firm sits in a different category?
Same shape of pattern, tuned for the specifics of your industry. You are reading the LinkedIn Ads troubleshooting hub.
Want someone to check yours? An ads audit is £250 + VAT per platform, back in 5 working days.