
To find business phone numbers that connect, source them in order: your own data first, public company sources second, provider data third. Aim for two numbers per contact, a mobile and a direct line, plus the company switchboard at account level. Verify before the first dial, then let call outcomes correct the rest.
Not every number is equal, and a CRM field called "phone" often hides which kind you have.
Label the type on every number you store. A mobile and a switchboard in the same field are two different strategies.
Stop at the first number that verifies for the right type. More lookups on the same contact add cost, not connects.
Company size changes the plan. In a ten-person company, the founder's mobile is the number, and the switchboard often rings straight to them. In a large group, the switchboard plus an extension, or a direct line inferred from the company's pattern, is frequently the realistic route. Match your effort to the account, not to a single rule for every contact.
A number in a database is a candidate, not a fact. Check three things before it reaches a rep.
Do not skip this because the list looks clean. A tidy spreadsheet is not a verified one, and the cheapest time to catch a bad number is before a rep spends a slot on it.
Then use the first calls as the real check. Log one outcome per dial: right person, wrong person, disconnected, switchboard or voicemail. A wrong-person result should pull the contact back for review, since the person may have moved.
We recommend two per contact: one mobile and one direct line, with the switchboard stored at account level. In our experience, a third personal number rarely adds connects, but it adds compliance risk and noise in the CRM.
The order of dialing matters too. From our analysis of 1,000,000 B2B cold calls in France, the average connect rate is 11.3%, and the best slots change with persona: founders pick up more around 10am, marketing around 4pm, HR around 1pm. Dial the mobile at the persona's best slot first, then try the direct line at another time of day. A second number at the same hour is a second chance to miss the same window.
A professional number is still personal data under GDPR. Document your legal basis, usually legitimate interest, inform people when you collect their details from a third party, and keep a suppression list that every tool respects. The rules for calling a professional differ from those for a consumer, and they vary by country, so ask counsel to confirm the position for each market you call.
We start with a different question: who deserves a number hunt in the first place? In our experience, reps use only about 20% of the signals already in their CRM, such as job changes, closed-lost deals and dormant leads. About 50% of Sonik's pipeline comes from signals that were already there.
So we put effort where there is a reason to call. Look for the number once the contact has a trigger, not before. It saves lookups and it makes the call better, since the rep opens with a reason. Our signals page explains how this works in the CRM.
Good phone data is a short, typed, verified list per contact, built from your own records first. Spend the effort on contacts with a reason to call. This week, take the 20 contacts you plan to call next, label every number you have for them as mobile, direct or switchboard, and see how many have no mobile at all.