The best phone number enrichment tool is the one that returns the most numbers that connect to the right person on your own list. Vendor match rates only tell you a number was returned, not that anyone picked up. The only test that counts is to dial every number and count the outcomes.

Why match rate is the wrong number to buy on

A match rate counts fields filled. It says nothing about whether the line is live, whether it belongs to the person you wanted, or whether it is a mobile, a desk line or a switchboard.

Two tools can show the same match rate and behave very differently once a rep dials. One returns mostly switchboards, the other mostly mobiles. One is strong on UK contacts and thin on France.

We analyzed 1,000,000 B2B cold calls in France with Quentin Despas (Katana). The average connect rate was 11.3%, and it moves between 11% and 20% depending on persona and time slot. Number quality is one input among several, which is why you need to measure it separately from timing and script.

The criteria that matter for France and Europe

Before you test anything, decide what you are scoring. These are the criteria we would write down first.

• Coverage by country. Ask for results split by market. A global average hides weak spots, and France, Germany and Spain rarely behave like the US.
• Number type. Does the tool label mobile, direct line and switchboard? A returned number with no type is hard to route.
• Precision. Of the numbers returned, how many reach the right person?
• Provenance and dates. Where did the number come from, and when was it last seen? You will need this for GDPR as well as for quality.
• Billing model. Do you pay per lookup, per number found, or per verified number? The cheapest headline price can be the most expensive per usable number.
• Write-back. Can results land in your CRM in separate fields, with source and timestamp?

A five-step test you can run in a week

1. Build a blind sample. Pull 200 contacts from your real target accounts, split across countries and seniority levels. Include 30 or 40 contacts where you already know the right number from past calls. That is your ground truth.
2. Send the same input to every tool. Same name, company and domain fields, same day. Do not let vendors pick the sample.
3. Dial everything that comes back. Use the same reps and the same time slots for every tool. Log one outcome per number: right person, wrong person, dead line, switchboard, voicemail.
4. Compute four figures per tool. Coverage (contacts with at least one number), precision (right person divided by numbers returned), no-number rate, and cost per right-person number.
5. Check the overlap. Look at which right-person numbers one tool found that another missed. This tells you whether a second source is worth paying for.

Be honest about sample size. With 200 contacts, a gap of a few points is noise. Treat it as a tie and decide on cost, number type and compliance instead.

Voicemail is not a failure here. A number that reaches the right person's voicemail is a good number, so log it as a good outcome. Dial each number more than once before you label it, because one unanswered call says almost nothing.

Do you need one provider or several?

In our experience, providers are uneven. One is better on mobiles for small companies, another on direct lines at larger ones, and the ranking changes by country.

The overlap check from step five answers the question with your own data. If the second tool adds a meaningful share of new right-person numbers, run them in sequence and stop at the first verified number. If it mostly repeats the first, you are paying twice for the same coverage.

Waterfalls add cost and complexity, so only build one where the test shows a gap you care about, usually in a single market or persona.

Compliance before you dial

Enriched numbers are personal data under GDPR, even when they are professional ones. Three things to settle with your DPO or counsel before volume goes up.

• Legal basis. B2B prospecting commonly relies on legitimate interest, which requires a documented balancing test.
• Information duty. When you obtain data from a third party, Article 14 asks you to inform the person, at the latest within one month or at first contact.
• Objection. Under Article 21, people can object to direct marketing at any time. Keep an internal suppression list and respect it everywhere.

Rules differ by country and by whether you call a professional or a consumer. France's Bloctel register, for instance, protects consumers, not professionals acting in their role. Ask a lawyer to confirm the position for each market you call.

How Sonik approaches phone data

Sonik is a dialer, not a data vendor. We see what happens after the number is returned, which is the part enrichment tools cannot see.

That is why we treat call outcomes as the final verification. Every dial tells you whether a number is dead, wrong or right. Parallel dialing gives reps about 5x more conversations, so a 200-contact test that takes a team days on a sequential dialer is a morning of work. You can learn more on the Sonik dialer page.

Our advice is to feed those outcomes back to your contact records, so the next enrichment decision is based on calls, not claims.

Conclusion

Phone enrichment tools should be judged on right-person numbers per euro, split by country, not on match rate. The test is cheap and the answer is specific to your list. This week, pull 50 contacts you already know the right number for, send them to the tools you are considering, and count how many come back correct.