What a match rate measures
The share of visits that can be matched to a company record from the network they came from. A visit from an office network or company VPN can usually be matched; a visit from home broadband or a phone usually can't.
What moves it
| Factor | Effect |
|---|---|
| Company size | Larger companies are easier to match |
| Device | Desktop matches better than mobile |
| Work pattern | Office days match better than home days |
| Region | Coverage of company IP data varies by country |
| Consent | Visitors who decline tracking aren't measured |
One vendor guide puts IP accuracy at 70–90% for companies over 500 people and 20–40% for companies under 50.
How to read results
- Count unmatched visits too. They're real demand.
- Judge quality, not volume: how many matched companies fit your profile.
- Check for false matches: internet providers, co-working spaces and hosting companies aren't buyers.
A quick way to estimate yours
| Step | Example (fictional) |
|---|---|
| Measured visits in a month | 2,000 |
| Visits matched to a company | 800 |
| Match rate | 40% |
| Unique companies | 590 |
| In your profile | 180 |
Watch the in-profile number; it's the one sales cares about.
Questions
Why do vendors quote higher match rates?
Definitions differ: some count matched sessions after removing mobile or bots, some count companies, not visits. Ask exactly what's counted.
Can we raise our match rate?
More desktop traffic and larger target companies raise it. Forms and logins add identified contacts on top.
Are false matches common?
Internet providers, co-working spaces and cloud hosts appear often. Filter them out before alerts.
Why match rates differ by market
Company identification relies on data that links network addresses to organisations. That data is richer for some countries and company types than others, and it changes as companies move to cloud services and remote work. A provider that performs well for US enterprises may perform differently for UK small businesses.
Test before committing
Run a provider on your own traffic for a month. Compare match rate, the share of matches in your profile and the number of obvious errors. The provider with the most matches isn't always the one with the most useful matches.
Start here: Company identification from AI. More articles in Company signals.