First, separate the match from the narrowing
Open the Matched Audience itself and read its status and its matched size, before any campaign is involved. If the list is still processing, stop: nothing about its size means anything yet. If it is ready and the matched size is far below the number of rows uploaded, the problem is in the file, and the first four rows of the table apply. Look at a sample of unmatched or doubtful rows and ask what they share: no domain, a group name, odd characters, personal addresses.
If the list matched well on its own, build the campaign audience from the list alone and read the forecast, then add each facet in turn: function, seniority, location, exclusions. The point at which the forecast drops below the documented minimum is the narrowing that needs attention. The account-based marketing guide explains why a tier-one list on its own is often too small by design; this page is about lists that should be large enough and are not.
Diagnosis table
Diagnosis
| Symptom | Check | Likely cause | Fix |
|---|---|---|---|
| Company list matches far fewer companies than rows | Share of rows with a website domain or page address | Company rows with names only | Add the domain and page address for every row |
| Matches include unexpected companies, or the main brand is missing | Rows using a group, subsidiary or trading name, or a generic word | Subsidiaries, trading names and common words matching the wrong page | Name the legal entity and its own domain; split groups into rows |
| Upload fails, or matches almost nothing from a clean-looking list | Column headers, file type and character encoding against the template | File not in the documented template or encoding | Rebuild the file from the current template and re-save it |
| Contact list matches a small share of a large file | Share of personal or long-unused addresses | Contact list of personal or old addresses | Use work addresses, add the other identifiers, or switch to a company list |
| List reads as tiny or empty shortly after upload | Processing status and upload time | Audience checked before processing finished | Wait for the ready status; do not re-upload |
| List alone forecasts well; campaign says too small | Forecast as each facet is added | Role filters narrowing the matched list below the minimum | Loosen the facet that removed most, or merge lists |
Fix, verify and prevent each cause
Company names with nothing to anchor them
A company name on its own is ambiguous: many firms share words, and LinkedIn has to guess which page you meant. The account-based marketing guide lists the fields LinkedIn uses to match a company page, including the website domain and the page address.
Fix: fill the domain for every row from the CRM or the company's own site, and add the LinkedIn page address for the accounts that matter most.
Verify: re-upload and compare the matched size with the previous version; the named target accounts now appear when you search the list.
Prevent: make domain a required field on account records in the CRM, so every future export carries it.
Names that point at the wrong page
Groups, subsidiaries and trading names confuse matching in both directions. A row for a group may match the holding company's page while the employees you want sit under the operating brand; a short or common name may match a different firm entirely.
Fix: list the operating companies as separate rows with their own domains, use the name each company uses on its page, and avoid rows that are a single generic word. The account-based marketing guide treats the account list as the programme's foundation, so the effort pays for itself.
Verify: spot-check the matched result for the target accounts and confirm that the expected operating brands are present.
Prevent: keep a parent and child structure in the CRM, and export the children for matching.
A file the uploader cannot read properly
The help centre documents the template for each list type, with fixed column headings and a file format. A file with renamed headings, merged columns, a spreadsheet format instead of the expected one, or an encoding that mangles accented characters can fail outright or match only the rows that survived.
Fix: download the current template from Campaign Manager, paste the data into it without changing the headings, save it in the documented format with a standard encoding, and upload again.
Verify: the upload completes without warnings and the matched size is in line with the number of complete rows.
Prevent: keep one export routine that writes straight into the template, rather than reformatting by hand each time.
Contacts whose addresses LinkedIn does not hold
A contact list can only match members whose accounts carry the address you uploaded. Event sign-ups, old trade show lists and personal addresses that people no longer use match poorly, and CRMs often hold the work address while members registered with a personal one. The targeting and Lead Gen Forms guide describes contact lists alongside the other Matched Audience types.
Fix: remove addresses that have bounced or gone unused, add the other identifiers the template accepts, and where reaching the account matters more than reaching the named person, switch to a company list with role filters.
Verify: the cleaned list matches a larger share of its rows, or the company-list alternative forecasts comfortably above the minimum.
Prevent: refresh contact lists from current CRM data rather than re-uploading old files.
A list read before it was ready
Matched Audiences take time to process after upload, as the account-based marketing guide notes. A list checked while still processing looks empty, and a nervous re-upload restarts the clock.
Fix: note the upload time, leave the list alone, and read it once its status shows ready.
Verify: the ready list shows a stable matched size on two checks a day apart.
Prevent: upload lists well before launch and record the ready date in the campaign plan.
Filters that cut a good list too far
A well-matched company list still has to pass every facet the campaign adds. Seniority, function, location and exclusions each remove members, and stacking all four on a modest list can take it below the documented minimum.
Fix: read the forecast as each facet is added and loosen the one that removed most, for example function and seniority instead of exact titles, as the targeting and Lead Gen Forms guide suggests, or combine tiers into one list.
Verify: the campaign forecast clears the minimum and the campaign starts serving.
Prevent: design audiences for lists that are known to be small, such as top-tier accounts, as one combined campaign from the outset.
Prevention: list hygiene before upload
Most low match rates are built in the spreadsheet, so a short hygiene pass before every upload prevents them. Check that every company row has a domain, that groups are split into operating companies, that the file uses the current template and encoding, and that contact rows carry current work addresses plus any extra identifiers available. Upload early and wait for the ready status. Then build the campaign audience in steps, reading the forecast at each facet, and record the matched size and the final forecast in the plan. Use the matched share as an operational indicator (KPIA metric tied directly to an objective.) for each refresh, so a sudden fall after an export change is noticed at once. Lists that later feed RemarketingAdvertising to previous visitors or engagers. of engaged members depend on this base, and a small, well-matched audience usually costs more per thousand impressions (CPMCost per 1,000 impressions.) than a broad one, which is acceptable when the accounts are the right ones. The LinkedIn business-to-business hub explains how list programmes are judged on account progress, and the LinkedIn troubleshooting hub covers campaigns that then fail to spend.
A logistics software list that matched a third of its accounts
A logistics software company uploads a company list of 600 target accounts for an account-based campaign. Once the list shows ready, it has matched 214 companies. A review of the file shows that 310 rows carry a company name only, 48 rows use a group name for businesses that trade under separate brands, and the file was saved from a spreadsheet with a renamed "Company" heading. The team adds domains from the CRM, splits the 48 group rows into 97 operating companies, and pastes everything into the current template. The re-uploaded list of 649 rows matches 571 companies. With function and seniority filters for operations and finance leaders, the campaign forecast now sits at about 52,000 members, where the earlier version had failed to clear the minimum once seniority was added.
Common mistakes
Uploading names without domains. Judging a list while it processes. Re-uploading to "speed things up". Matching group names and wondering why the operating brands are missing. Blaming the file when the campaign filters did the narrowing.
