A commercial cleaning company has been running LinkedIn ads for two months. It sells nightly office cleaning to facility managers and property managers in one metro area, and it has spent about $6,000 so far. The account shows eight leads. None has turned into a signed contract. The owner has a new headline written and is about to swap it in, on the theory that the ad isn't working.
That is the last thing to change, not the first. When LinkedIn ads aren't working, the cause is usually who saw them, not what they said. You can't judge an ad until you know where it ran and who it reached, and LinkedIn gives you a report for each of those questions. Most owners never open them. They go straight to the creative because the creative is the part they can see.
This is for an account that is already running. If you are still building one, the settings to get right before launch are in LinkedIn Ads Aren't Expensive. Your Setup Is. What follows is the order to check things in, with the report that answers each question and what you do when the answer is bad. Work through it top to bottom. Each check can explain a dead account on its own, and fixing a later one while an earlier one is still broken just wastes another month.
A New Headline Can't Fix an Ad Shown to the Wrong People
Rewriting the ad feels productive. It is also the one change that can't tell you anything while the rest of the account is broken. If half your budget is being shown to the wrong people, a better headline gets shown to the wrong people too, and you conclude that the new headline failed as well.
Agencies that audit LinkedIn accounts describe the same pattern. A business arrives saying its ads aren't working. It has already tested a few headlines and found no difference between them. Then the agency opens the account and finds that nearly all of the spend went somewhere other than the LinkedIn feed. The headline tests were real, but they were run on an audience nobody chose, so they could not have shown anything.
Our cleaning company is in the same spot. Before the new headline goes in, it needs to know what its $6,000 actually bought.
Audience Network Sends Part of Your Budget Off LinkedIn
LinkedIn can show your Sponsored Content ads in two places: in the LinkedIn feed, and on the LinkedIn Audience Network, which is a set of partner apps and websites outside LinkedIn. Those clicks are cheaper, and that is the problem. You are paying for LinkedIn's knowledge of who someone is, and off the platform that knowledge gets thinner. LinkedIn's own help page on key results says some Audience Network ads use a model to deliver and measure, so some results "might come from audiences that sometimes align only partially with your targeting criteria."
Finding out takes one click. In Campaign Manager, open the Advertise page, pick your campaigns or ad sets, then open the Breakdown menu and choose On/Off Network. Every row splits into "On LinkedIn" and "Audience Network," with impressions, clicks and spend for each. If most of the spend sits in the second row, you have found your first problem, and every other number in the account is mixed with it. We made the full case for switching it off in Should You Turn Off LinkedIn Audience Network?
The second setting in this check is audience expansion. It shows your ads to people LinkedIn considers similar to your audience, and LinkedIn's help page says it switches expansion on by itself on every ad set where the option exists, so you have to clear it. We covered the other automatic settings in LinkedIn Took Away the AI Campaign. It Kept the AI. Two details make it hard to spot. The audience size in the forecast panel does not include the people expansion adds, so the audience you approved looks smaller than the one you are buying. And your reports mix both groups together, so nothing in the main table marks which impressions came from expansion.
For the cleaning company, the fix costs nothing: clear both checkboxes in each ad set and note the date. Every number from that day on describes the audience it actually chose.
The Biggest Employers in Town Soak Up Your Impressions Without Ever Buying
Once you know the ads ran on LinkedIn, the next question is whether they reached the people you meant. Click Professional Demographics on the same Advertise page. LinkedIn shows impressions, clicks and spend broken down by company, job title, job function, seniority, company size, industry and location, at the account, campaign, ad set or ad level.
Read it with two questions in mind.
Which companies are taking the budget? Big employers have more people who match your targeting, so they collect more impressions. Our cleaning company targeted facility managers in its metro area and found two hospital systems and a university near the top of the company list. None of them buys office cleaning from a 30-person company. They were simply the biggest employers in town with people called facility managers. We covered how to cap the companies that hog delivery in A Handful of Companies on Your LinkedIn List Get Most of Your Ads.
Which job titles actually saw it? LinkedIn groups related titles, so a target of "facilities manager" can quietly include maintenance supervisors and building engineers who never sign a cleaning contract. If titles you would never call show up with real spend, add them as exclusions. LinkedIn Tells You Exactly Who Clicked. Almost Nobody Looks. walks through reading this report in full.
Two limits are worth knowing so a blank report doesn't throw you. LinkedIn says demographic data can take 24 to 48 hours to appear, and it hides any slice that falls below a minimum level of delivery, to protect members' privacy. A small campaign may show only a few rows. Widen the date range before deciding the report is empty.
On an Engagement Campaign, LinkedIn Counts a Like as a Click
This is the check most owners have never run, because the number they watch looks fine.
The Clicks column in Campaign Manager counts what LinkedIn calls chargeable clicks, and what counts as a click changes with your campaign objective. On an engagement campaign, LinkedIn's help page lists clicks to your landing page, clicks to your LinkedIn Page, social actions (likes, comments and shares), clicks on the line showing who reacted, and Page follows. Every one of those counts as a click, and every one feeds your click-through rate. Only one of them puts a person on your website.
The number that does is Landing page clicks. Add it from the Columns menu and set it beside Clicks. On a website visits campaign the two sit closer together, because LinkedIn does not charge for clicks on your company name or for expanding the intro text. On an engagement or brand campaign, the gap can be wide, and that gap is the difference between how busy the ad looks and how much traffic it actually sent you.
Then check the landing page clicks against your own website analytics for the same days. If LinkedIn says 140 people clicked through and your site saw 60 visits from LinkedIn, the page is slow, the link is broken, or people are leaving before it loads. No headline fixes that.
The cleaning company's campaign used the engagement objective, because it was cheaper per click. Its click-through rate looked healthy. Its landing page clicks were less than half of its clicks. A good share of what it had been calling traffic was people liking the post.
LinkedIn Can't Credit a Lead Its Tag Never Saw
Sometimes the campaign is producing leads and the account just isn't recording them. The symptom is easy to spot once you look for it: Campaign Manager shows zero or close to zero conversions, while your inbox or CRM has inquiries that mention LinkedIn.
The usual causes are plain. The Insight Tag is missing from the page where the form lives, or from the thank-you page. The conversion is set to fire on the wrong page. Or the form sits on a separate booking tool where the tag was never added. You can find the tag under Measure, then Signals manager, in Campaign Manager, and LinkedIn shows whether it has seen recent activity on your site.
Counting also has a clock. LinkedIn's website tag only credits a conversion inside its window, and on a long sale that window closes before many buyers act, which we covered in Your LinkedIn Website Tag Stops Counting at 90 Days. If your buyers take months, judge the channel by your CRM, not by the conversion column, as LinkedIn Isn't Failing. You're Measuring It Wrong. lays out.
If the counting is broken, every decision you have made from the account was made on bad numbers. Fix it before judging anything else, and expect your cost per lead to look different, possibly better, once real leads start showing up in the column.
Three Visits a Day per Campaign Is Too Few to Judge Any Ad
Some accounts aren't failing. They just haven't bought enough clicks to say anything yet.
Do the arithmetic per campaign, not per account. Take the monthly budget, divide by 30, divide by the number of campaigns, then divide by what a landing page click actually costs you. Our cleaning company spends $3,000 a month across three campaigns. At about $12 per landing page click, each campaign buys fewer than three visits a day, or about 80 a month. Its eight leads in two months work out to a form rate under 2%, which is a little over one lead per campaign per month. One lucky week doubles that number and one quiet week erases it. Nobody can tell whether a campaign works from that.
The fix is fewer campaigns, not more money. Put the budget behind one audience and one offer until it buys enough clicks a day to read. You Have Four Campaigns and a Budget for One. has the full math and the order to merge things in.
A Free Quote Asks a Stranger for a Meeting Before They Know Your Name
Now look at the eight leads. Leads that never become sales usually trace back to the offer, not the targeting.
The cleaning company's ad offered a free walkthrough and quote. That is a meeting. It asks a facility manager who has never heard of the company to schedule time with a salesperson. The people who say yes to that from a cold feed ad tend to be the ones who say yes to every free quote, which is why the leads arrived and the contracts didn't. The managers it actually wants were happy with their current cleaner that month and scrolled past.
Read the form answers and the sales notes for every lead and sort them into three groups: the right company and the right person, the right company and the wrong person, and the wrong company. Then match them to the offer. If most leads are wrong-company, go back to the companies check. If they're the right people but they don't book, the ask is too big for a first touch. Something smaller, like a cleaning cost checklist or a one-page comparison of in-house and contract cleaning, gets the right person to raise a hand without committing to a sales call. Nobody on LinkedIn Wants a Demo. Offer Them Something Smaller. covers how to build that first offer.
You Can't Call the Ad Weak Until Delivery, Tracking and the Offer Hold Up
If the money ran on LinkedIn, reached the companies and titles you meant, sent real visits to your site, got counted, bought enough clicks to read, and the offer fits a first touch, then the ad itself is fair game. Now a creative test can tell you something.
A real creative problem has a recognizable shape. Delivery is clean, the demographic report shows the right people, and the landing page click rate is still low compared with your own other ads to the same audience. That last comparison matters. LinkedIn's averages mix every industry and objective together, so measure against your own account. When that is the picture, the ad is probably talking about your company instead of the reader's problem, and You Fixed the Targeting. The Ad Still Says Nothing. is the place to start.
For the cleaning company, the order mattered. Clearing Audience Network and audience expansion, excluding the hospitals and the university, moving to a website visits campaign and collapsing three campaigns into one changed more in a month than any headline would have. The new headline is still in a document. It may get tested next month, on an audience that can actually judge it.
Half an Hour a Month in Campaign Manager Beats Another Headline Rewrite
None of this needs a new tool. It takes about half an hour in Campaign Manager once a month, and the order is the point:
- Where the money ran: On/Off Network breakdown, plus audience expansion in each ad set.
- Who it reached: Professional Demographics by company and job title.
- Whether clicks became visits: Landing page clicks against Clicks, then against your website analytics.
- Whether leads get counted: Conversions in Campaign Manager against the inquiries in your CRM.
- Whether the budget can answer: landing page clicks per campaign per day.
- Whether leads can buy: a review of every lead against the offer.
Then, and only then, the ad.
When we take over a LinkedIn account, the first three checks are where we start, because they cost nothing to fix and they change what every other number means. If you would rather have someone run this every month and fix what it turns up, that is the work our LinkedIn ads management team does. If you would rather do it yourself, you now have the order, and the first check takes one click.
