Data coverage is how many of the companies you care about a dataset actually contains, and how fully each record is filled in. It is always measured against a defined list, such as your customer list, a set of listed companies or the firms in one country, and never in the abstract.
Measuring it
Take your own list, match it to the dataset on a stable key such as domain, ISIN or company ID, and divide: matched records over list records is the match rate. Then compute a fill rate for each field you need among the matches, and split both by country, size and industry, because an average hides the segment you care about.
Acme Robotics (illustrative) matches 3,200 of its 4,000 CRM accounts, an 80 percent match rate, and 2,400 of those 3,200 have hiring data, a 75 percent fill rate.
Why counts mislead
A vendor's headline count says how much it holds, not how many of the companies you track it holds. A count that grows over time may reflect new coverage and not new activity in the market: if a dataset adds companies every day, a raw count of events rises by itself. Same-store cohorts fix this by comparing only the companies covered for the whole period. Coverage also differs by dataset within one vendor, since a company can be present in one and absent from another.
Coverage in Fokals data
The company index holds listed companies worldwide (equity listings in 79 countries), the brands they own and verified private companies, each with one stable company ID. Coverage is stated per dataset. Technology Stack and Website Profile describe the websites of companies in the index. Job Postings and Hiring Activity follow every role a company publishes. Company News carries company announcements and regulatory disclosures, Employee Headcount carries stated headcount over time, and Web Traffic carries the monthly traffic tier of each website. Sample data for the companies you track shows the match rate and fill rate on your list before you license.
The coverage page states the scope, and the methodology describes how each dataset is built. The market series use same-store cohorts, so growth in the index does not read as growth in the market.
Related terms
Coverage is judged beside data freshness. It is measured by domain matching or entity resolution, and a same-store cohort keeps a growing index from distorting a trend.
Frequently asked questions
What is the difference between data coverage and data completeness?
Coverage asks how many of the entities you care about are in the dataset. Completeness asks how many of the fields you need are filled for the entities that are. A dataset can cover every company on a list and still be incomplete if most of its rows lack the field you want, or be complete for a narrow slice and cover little of your market. Measure both, and report them by segment.
What is a good match rate for company data?
No threshold is good in itself. It depends on the companies and the use: a list of listed companies usually matches more fully than a list of small private firms, and a prospecting list tolerates gaps that a portfolio-monitoring list does not. Measure the match rate on your own list, by segment, and judge it against what the decision needs rather than against a figure the vendor quotes for its whole index.
How do you check data coverage by country?
Take your own list and split it by country, size band and industry. Compute the match rate and the fill rate of each field you need inside each segment, and compare them with the decision the data feeds. Coverage follows what companies publish, so it differs by field and by market. Ask a vendor for coverage by country for the fields you need, not one global figure, and test a sample on the companies you track.