LinkedIn Matched Audiences: Why They Underperform and How to Fix Them

LinkedIn Matched Audiences let you target people and companies you already know: your website visitors, your CRM contacts, your target account list. Building one takes a few minutes. The trouble usually starts after the upload, when a contact list comes back with a low match rate, or a website audience sits gray under 300 members for weeks.

This guide covers each type of Matched Audience, how LinkedIn calculates match rate, and the setup choices that decide whether an audience is usable. It assumes you already run campaigns, so it skips the basics of LinkedIn ads optimization.

What Are LinkedIn Matched Audiences?

Matched Audiences is LinkedIn’s name for targeting built from your own data. LinkedIn targeting by attribute (job title, industry, company size, skills) describes the kind of person you want to reach. A Matched Audience names them: these companies, these contacts, the people who visited this page. Some marketers call them LinkedIn custom audiences, borrowing Meta’s term, but in Campaign Manager the feature is Matched Audiences.

According to LinkedIn’s setup guide, you can create one in three ways:

• Upload a list of contacts or companies as a CSV.
• Retarget people who visited your website or engaged with your ads, Lead Gen Forms, Company Page or events.
• Connect a marketing partner, such as your CRM or a data provider, and let it send segments to Campaign Manager.

Whatever the source, LinkedIn matches it against member accounts (or LinkedIn Pages, for company lists), and only the matched part becomes your audience. Members who’ve opted out of ad targeting aren’t included, and in the EEA and Switzerland LinkedIn only matches members who’ve opted in. Every Matched Audience needs at least 300 matched member accounts before an ad set can use it.

Types of LinkedIn Matched Audiences

Each type matches against something different, which is why the same set of accounts can produce very different audience sizes depending on how you bring it in.

Website retargeting audiences

Website audiences come from the LinkedIn Insight Tag. Once it’s on your site, LinkedIn builds the audience from visitors who are LinkedIn members, filtered by the pages they visited and how long ago.

The audience starts building on the day you create it and doesn’t include earlier traffic, even if the tag has been on your site for a year, so create the website audiences you’ll need well before the campaign. Only the share of traffic LinkedIn can match to members counts, and LinkedIn suggests giving a new website audience 90 days to build up.

Website audiences are the core of LinkedIn retargeting: demo-page visitors who didn’t book, pricing-page visitors, readers of one specific product page.

Contact list audiences

A contact list is a CSV of people, matched mainly by email address. LinkedIn also accepts first and last name (more accurate with the company name added) and mobile device IDs. Emails can be plain or hashed with SHA-256.

One detail explains most low contact match rates: LinkedIn matches only against verified email addresses on member accounts. Plenty of members verified a personal address when they signed up, while your CRM holds their work email. That’s why LinkedIn recommends uploading at least 10,000 personal or professional emails to end up with 300 verified matches, and why it helps to include both addresses for a contact when you have them.

Contact lists also can’t be combined with member interests and traits in the same ad set. Layer them with job function or seniority instead.

Company list audiences

A company list matches your accounts to LinkedIn Pages, then reaches the members who work at those companies. That makes it the backbone of account-based marketing on LinkedIn. Each row needs at least one of company name, website, email domain, LinkedIn Page URL or stock symbol, and the more fields you fill in, the better the match.

The Page URL is the field that matters most. LinkedIn’s troubleshooting guide calls it the most efficient way to match, as long as the URL ends in the company’s name (linkedin.com/company/name) and not a number ID. A company name on its own is the least reliable input, because several businesses can share a name and LinkedIn has to pick one.

LinkedIn recommends 1,000 or more companies, and the file needs at least 300 rows to upload, so a list of 80 tier-one accounts won’t upload by itself. Tier it inside a larger company list, or use company-name targeting for the smallest tier. You can also save accounts already engaging with you as a static or dynamic list from Companies (Plan → Companies).

Account-based audiences from your CRM

If your account list lives in a CRM, a marketing partner integration can send it to Campaign Manager instead of a CSV, and keep sending updates as the CRM changes. Processing averages 48 hours or less, but a large list can take up to a week and may arrive in portions.

Expect the audience count to come in below the segment size, since one member can have several verified emails and two CRM rows can be the same person.

Predictive audiences (formerly lookalike audiences)

LinkedIn discontinued lookalike audiences on February 29, 2024, and predictive audiences replaced them. You pick one source type (a contact list, company list, conversion, Lead Gen Form or retargeting audience) with at least 300 members, and LinkedIn’s model finds members predicted to act like the people in it.

Predictive audiences refresh daily, so they follow your source as it changes. You set the size with a slider, and once the audience shows Ready it can take up to five days to reach that size. A company or contact list seed can end up inside the result, so exclude the list if you only want new people. LinkedIn’s guidance is to match the source to the objective: Lead Gen Form submissions for lead gen campaigns, conversions for website conversion campaigns, closed-won contacts when you want more people like your customers.

How LinkedIn Matched Audiences Match Rates Work

Match rate is the share of entries in your source data that matched to member accounts, or to LinkedIn Pages for company lists, within the past seven days. It’s calculated on the last list you uploaded or updated, and it sits in the Match rate column on your Audiences page.

LinkedIn doesn’t publish a typical match rate, so there’s no benchmark to hold yours against. What you can do is read the two numbers on the Audiences page together:

• Match rate tells you how well your data matched.
• Last audience count tells you how many reachable members you have, after duplicates and opted-out members are removed.

They often disagree. LinkedIn’s own example: a 2,000-row list where 1,400 rows match, but 200 of those are duplicates, leaves an audience of 1,200.

One display rule is worth knowing before you start troubleshooting: if fewer than 300 member accounts matched, the match rate shows as <5% whatever the real figure is, the audience count shows < 300 members, and the list name turns gray.

What lowers a match rate

Most low match rates trace back to the file:

• Edited template headers or the wrong encoding. Renamed or removed headers make the upload fail. If the file has special characters, save it as UTF-8.
• Work emails only. Members whose verified address is personal won’t match on a work email.
• Company rows with a name and nothing else. Add the website and Page URL.
• Page URLs with a number ID instead of the company’s name.
• Broken hashing. Hidden characters in a hashed email list, or a hash other than SHA-256, will stop rows matching.
• Stale data. People who’ve left the company your list says they work at.

For company lists, LinkedIn suggests going through the companies that didn’t match, adding their Page URLs, and re-uploading.

How to Upload and Use a Contact or Company List

1. Get the template. Download the company or contact template from the upload dialog in step 3. Add your rows below the headers, keep it between 300 and 300,000 rows and under 20 MB, and save it as a CSV.
2. Start a new audience. Go to Plan → Audiences, click Create audience and select Matched Audience.
3. Upload the list. Select Company / Contact, click Next, name the audience, choose the list type, click Select list and upload your file. Click Agree & Upload.
4. Wait for matching. The status shows Building for up to 48 hours (rarely 72). After 72 hours, check the list requirements and re-upload it as a new audience.
5. Check the result. Read match rate and last audience count together. A Failed status means the list processed but matched fewer than 300 members, so add more companies or contacts.
6. Apply it to an ad set. Add the audience, refine it with job function or seniority, and exclude current customers. After you apply a processed list, allow another 24 hours before it shows up in delivery.

Website Retargeting as a Matched Audience

Website audiences don’t need a file. They need the Insight Tag on every page you want to retarget from.

To install the tag, go to Measure → Signals manager → Insight Tag in Campaign Manager. You can install it yourself, send it to a developer or use a tag manager. The code goes right before the closing body tag in your site’s global footer. LinkedIn validates the domain once a tagged page is visited, which takes anywhere from a few minutes to 24 hours.

Then build the audience:

1. Go to Plan → Audiences → Create audience → Matched Audience.
2. Select Website, click Next and name the audience.
3. Under When did they engage?, choose the lookback window.
4. Pick website actions LinkedIn has already collected (button clicks, page visits, form submissions), or click Manually enter page URLs and set a rule:

– Starts with for every page under a path, including URLs with dynamic parameters

– Equals for one exact page

– Contains for pages with a keyword in the URL

5. Use + And also to narrow the rule (a URL that starts with example.com and contains thank-you), or + Add a new OR rule to build from several pages.
6. Click Agree & create.

When you enter URLs, leave out http:// and https://, and include “www” only if visitors see it in the address bar.

The status should change to Sending signal within 24 hours. No signal usually means the tag isn’t on the pages your rule points to. If the audience is stuck under 300, switch an Equals rule to Starts with, choose a longer lookback window, or add several website audiences to the same ad set, since segments of the same type add up.

LinkedIn Matched Audiences Best Practices

• Lists that stay current. Re-upload contact and company lists as your CRM changes, or connect a marketing partner so updates flow through. An audience that hasn’t been updated or used in an active or draft ad set for 30 days is archived, and at 90 days it expires and has to be re-uploaded.
• Demographic layers for precision. Add job function or seniority to a company list so the ads reach the decision makers at each account.
• Exclusions on every prospecting campaign. Exclude current customers and accounts with an open deal, or budget goes to people who’ve already bought or are already talking to sales.
• Separate campaigns for warm and cold audiences. A pricing-page visitor and a cold target account need different messages.
• Bids that fit a small audience. A short list of high-value accounts often needs a more competitive bid to deliver at all. For warm audiences, test CPM bidding against CPC, since being seen by the right account can matter more than the click.
• The right seed for predictive audiences. Start from the audience that already converts for that objective, even if it’s smaller than your biggest list.

Common LinkedIn Matched Audiences Mistakes

• Launching before the audience is Ready. An ad set starts delivering once a building audience reaches 300, so the first days run on a partial audience and give you a misleading read.
• Uploading a full CRM export. Most of those contacts aren’t in-market right now. Filter to recent engagement, open opportunities or the accounts sales is working.
• Mixing audience types in one ad set without meaning to. LinkedIn treats different Matched Audience types as AND: include a website audience and a company list, and members have to be in both. Segments of the same type add up instead. Location is required as well, which narrows things further. LinkedIn suggests an ad set audience of at least 50,000 members to drive results, so a narrow intersection rarely has room to deliver.

• Running the same audience in too many campaigns at once. Frequency climbs and the creative wears out faster. Watch frequency in reporting and rotate creative.
• Accepting a low match rate without asking why. Check headers, verified emails, Page URLs and how fresh the data is before you write the list off.

Building Higher-Intent Account Lists

A Matched Audience is only as good as the list behind it. A CRM export holds everyone you’ve ever spoken to: churned customers, deals lost two years ago, contacts who have since changed jobs. Upload it as it is and LinkedIn will faithfully match a lot of people who aren’t buying.

A sharper input is a list of accounts showing intent right now, like companies visiting your pricing and product pages or coming back more than once in a week.

Website visitor identification is how you get that list. DemandSense shows which companies and people are on your site and how each one fits your ICP, and those accounts make a sharper Matched Audience seed than a full CRM export. Add B2B intent data from outside your site, and your company list reflects what accounts are doing this month.