# Phone number API: mobiles and direct dials for B2B contacts

> How to add phone numbers to B2B contacts programmatically, when a call beats an email, and how to keep dialing lists clean and compliant.

Source: https://www.b2benrich.com/guides/phone-number-api  
Author: Dag Holmen, B2Benrich · Updated 2026-10-28

A phone number API adds mobile numbers and direct dials to B2B contacts you've already found. You pass in a contact and get back the numbers that reach that person, ready for a dialer or a rep's call list. B2Benrich returns both mobiles and direct dials through the same API you use for people and company data.

Getting the numbers is the quick part. This guide covers the rest: one storage format, a known type for every number, a check before anyone dials, and the calling rules where the person is.

## When a call beats an email

Calls cost rep time, so they belong on a shortlist. They pay off when:

- **Timing matters.** A target account just requested a demo or started a trial.
- **Email has gone quiet.** No reply doesn't always mean no interest. A call is a different door.
- **The role doesn't live in an inbox.** Operations leads, field managers and founders of small companies are often easier to reach by phone.
- **The deal is big enough.** For a few high-value accounts, a five-minute conversation beats another sequence step.

For large, cold lists, email and ads reach more people per hour of work.

## Mobile vs direct dial

| Type | What it reaches | Good for | Watch for |
|---|---|---|---|
| Mobile | The person's cell, wherever they are | Remote and hybrid workers, field roles | Often doubles as a personal phone. Stricter US rules for autodialed calls and texts |
| Direct dial | A desk line that rings the person, past the switchboard | Office-based roles | Goes stale when people move desks or jobs. Often forwards to voicemail |
| Company main line | Reception | Asking for someone by name | Doesn't reach a specific person |

Store the type with every number, because reps and compliance rules both care. One catch: in the US and Canada, mobile and landline numbers look the same, so you can't tell the type from the digits. Keep the type your data gives you, and treat a number of unknown type as a mobile.

## Build the call list

A cold calling list starts with named people. The [prospecting API](https://www.b2benrich.com/guides/prospecting-api) returns person records filtered by title, level and company, each with an `id`, `linkedin_url`, `company_domain` and `country_code`. If you already have LinkedIn URLs, the [people enrichment API](https://www.b2benrich.com/guides/people-enrichment-api) gets you the same kind of record.

```bash
curl --request POST \
  --url https://api.b2benrich.com/v1/search/prospector \
  --header 'Content-Type: application/json' \
  --header 'X-Api-Key: YOUR_API_KEY' \
  --data '{
    "contact_filters": { "job_function": "Operations", "job_level": "Director" },
    "company_filters": { "employee_count_range": "51-200", "hq_country": "United States" },
    "page": 1,
    "page_size": 25
  }'
```

Check allowed filter values with `POST /v1/search/meta/job_level` and its siblings before you hard-code them.

## Add phone numbers to each contact

Next, request phones for each person on the list.

The script keeps the phone lookup behind one placeholder function. Everything around it (normalizing, deduping, tagging the type) works whatever the response looks like. It uses `phonenumbers`, the Python port of Google's libphonenumber.

```python
import phonenumbers

def get_phone(person):
    # TODO: replace with the B2Benrich phone endpoint.
    # Map its response to a list like [{"raw": "+1 512 555 0147", "type": "mobile"}].
    raise NotImplementedError

def to_e164(raw, region):
    try:
        number = phonenumbers.parse(raw, region)
    except phonenumbers.NumberParseException:
        return None
    if not phonenumbers.is_valid_number(number):
        return None
    return phonenumbers.format_number(number, phonenumbers.PhoneNumberFormat.E164)

def add_phones(person):
    # Two-letter region, used only for numbers written without a +country code
    region = person.get("country_code") or "US"
    phones, seen = [], set()
    for p in get_phone(person) or []:
        e164 = to_e164(p["raw"], region)
        if e164 and e164 not in seen:
            seen.add(e164)
            phones.append({"number": e164, "type": p.get("type") or "unknown"})
    return {**person, "phones": phones}
```

## Normalize every number to E.164

E.164 is the international format: a plus sign, the country code and the number, with no spaces or punctuation. `(512) 555-0147` in Austin becomes `+15125550147`. `020 7946 0018` in London becomes `+442079460018`.

Stored this way, the same number written three ways dedupes to one value, dialers and SMS tools get the country code they expect, and nobody guesses which country a ten-digit number belongs to. Parse with a library, not a regex, and format numbers for humans only on display.

## Validate before numbers reach the dialer

A format check catches numbers that can't exist. It can't tell you whether a valid-looking number is still connected, or still belongs to the person. B2Benrich's docs list a phone validation endpoint, `POST /v1/validate/phone`, covered alongside email checks in the [email verification API guide](https://www.b2benrich.com/guides/email-verification-api).

Your reps are the other check. Log a disposition on every call: connected, voicemail, wrong person, not in service. Drop dead numbers and send wrong-person numbers back for re-enrichment.

## Compliance basics before anyone dials

> This is a list of things to check, not legal advice. Calling rules depend on where the person you call is, so confirm the details with counsel for each country you dial into.

- **Keep your own do-not-call list.** When someone asks not to be called, add them that day. Check the list before every calling block.
- **Screen against national registries.** In the US, that's the National Do Not Call Registry. Business calls are treated differently from consumer calls, but a mobile on a B2B list may also be someone's personal phone. In the UK, screen against the TPS and the Corporate TPS before live calls.
- **Be careful with US mobiles.** Under the TCPA, calls to a mobile made with an autodialer or a prerecorded voice need the person's prior express consent, and prior express written consent for telemarketing. Texts count as calls. Ask counsel whether your dialer counts as an autodialer.
- **Check consent rules by country.** In parts of the EU, marketing calls need some form of consent even when the number belongs to a business.
- **Call in their working hours,** in their time zone, not yours.
- **Record where each number came from,** with a date, so you can answer anyone who asks.

## Keep numbers fresh

Direct dials go stale when people change jobs. Mobiles usually move with the person, so the number stays live while the job changes. Re-enrich before each calling block, and catch moves with [job change tracking](https://www.b2benrich.com/guides/job-change-tracking). B2Benrich is one plan, unlimited API usage for $500 a month, so re-checking the whole list costs the same as checking ten numbers.

Pair every phone with a work email from the [email finder API](https://www.b2benrich.com/guides/email-finder-api), so a rep can follow a call with a note.

## Frequently asked questions

### What is a phone number API?

An API that returns phone numbers for a contact you send it. For B2B, that usually means a person’s mobile and direct dial, so reps and dialers can reach them without searching by hand. B2Benrich returns both through the same API as its people and company data.

### What’s the difference between a mobile number and a direct dial?

A mobile is the person’s cell phone and reaches them wherever they are. A direct dial is a desk line that rings the person without going through the switchboard. Store the type with each number, because reps and calling rules treat them differently.

### What format should I store phone numbers in?

E.164: a plus sign, the country code and the number, with no spaces, like +15125550147. It makes deduping reliable and it’s the format dialers and SMS tools expect. Parse numbers with a library such as libphonenumber rather than a regex.

### Can I tell if a number is a mobile from the digits?

Not in the US or Canada, where mobile and landline numbers use the same format. Many other countries give mobiles their own prefixes. Keep the type your data source gives you, and treat numbers of unknown type as mobiles.

### Can I cold call business mobiles in the US?

This isn’t legal advice. Live, hand-dialed calls are treated differently from autodialed ones. Under the TCPA, autodialed or prerecorded calls and texts to a mobile need prior express consent, and written consent for telemarketing. Keep a do-not-call list and check the details with counsel.


## Sources

- [ITU-T — Recommendation E.164: The international public telecommunication numbering plan (2010)](https://www.itu.int/rec/T-REC-E.164)
- [Cornell Law School LII — 47 CFR § 64.1200: Delivery restrictions](https://www.law.cornell.edu/cfr/text/47/64.1200)
- [FTC — National Do Not Call Registry](https://www.donotcall.gov/)
