A matched audience is an advertising audience built from a list you already own - accounts, contacts, customers, leads - which an ad platform matches against its own members so you can target them, exclude them, or use them as a seed for lookalike expansion. LinkedIn calls the product Matched Audiences; Meta calls it Custom Audiences from a customer list; Google calls the mechanism Customer Match; Reddit, X, TikTok and Pinterest offer the same idea under their own names. The mechanism is identical: identifiers in, matched members out, individuals never revealed.
The definition
You upload a file of identifiers, or push them through the platform's API from a connected account. The platform hashes the values (or accepts them pre-hashed), compares them to its member records, and creates an audience containing the members it recognized. You learn the audience size and, on most platforms, the match rate, and never which individuals matched. From there the audience behaves like any other targeting option, and it can be excluded from a campaign as easily as included in one. The full operating manual - the pipeline from rows to a live audience, hygiene, refresh cadence and the approval gate in front of the ad account - is AI agents for matched and custom audiences.
What the platforms match on
| Platform | Match keys | Notable |
|---|---|---|
| Company name, domain, page URL; email, name, title, company | Matches at the company level as well as the contact level | |
| Meta | Hashed email, phone, name, location fields, mobile advertiser ID | More identifiers per row raise the match |
| Hashed email, phone, name with country and ZIP, device ID | Policy-gated; a documented minimum before serving | |
| Reddit, X, TikTok, Pinterest | Hashed email, mobile advertiser ID; X also accepts handles | Mostly retargeting and suppression for these lists |
The consequence for B2B is direct: LinkedIn is where account lists belong, because it is the platform that matches companies, so an ICP list becomes an audience without personal identifiers. Everywhere else the row needs personal identifiers, which makes the person-data half of the list - the emails and phones found and verified upstream - the limiting factor. Google's version has its own entry, what is Customer Match.
The four audience types
Retargeting audiences hold people already in a conversation with you - contacts in sequences, open opportunities, signal movers - so paid media stays visible alongside outbound. Suppression audiences hold people you must not pay to reach - customers, active opportunities, competitors, employees, opt-outs - and are excluded from campaigns and removed from lookalike seeds. ABM account lists hold the ICP or a signal-qualified segment, targeted at the company level on LinkedIn and at the contact level elsewhere. Lookalike seeds hold closed-won customers and best-fit accounts for the platform's own expansion, and are only as good as the seed and its suppressions.
The match rate
The match rate - matched members divided by uploaded rows - is a data-quality score in disguise. Verified business emails match; guessed patterns and personal addresses on a B2B list mostly do not. Multiple identifiers per row give the platform more than one chance to recognize a member. Formatting drift before hashing - a stray capital, a trailing space, a phone without its country code - hashes to a different value and matches nothing. Reading the rate per list version turns the audience program into a feedback loop for the list program, and it is why AstroFabric's audience agent builds from a verified, persistent list, applies the suppression lists first, and stages the push for approval with the membership count and match rate recorded against the list version.
Frequently asked questions
What is a matched audience?
An ad audience built from a list you own: you upload identifiers such as hashed emails, phones, names or company domains, the platform matches them against its members, and the recognized members become an audience you can target, exclude or use as a lookalike seed.
What is the difference between a matched audience and a custom audience?
Mostly naming. LinkedIn calls the product Matched Audiences, Meta calls it Custom Audiences from a customer list, Google calls the mechanism Customer Match. All of them match uploaded identifiers against platform members; the formats, keys and minimum sizes differ.
Can a matched audience be built from company names?
On LinkedIn, yes: company lists match on name, domain and page URL, so an account list becomes an audience without personal identifiers. Other major platforms match on personal identifiers, so the list needs verified emails or phones per row.
Why is my match rate low?
Usually the list: unverified or personal emails on a B2B upload, single-identifier rows, formatting drift before hashing, or a segment the platform’s membership does not cover. Verified emails, multiple identifiers and clean normalization move the rate more than anything on the platform side.
Sources
Every playbook on this blog ships as a runnable mission.
Open a workspace and the playbook library is waiting - describe the outcome and the agents carry it end to end, on your plan's monthly credits.