
Buy a B2B phone list when you need to enter a new segment quickly and the seller lets you test a sample on real calls. Build it yourself when the accounts are already in your CRM, which is more often than teams think. In both cases, judge the list on connect rate and meetings booked, never on the number of records.
A list of 20,000 numbers that nobody picks up is not an asset. It is a month of your SDRs' time.
Before you compare vendors, open your CRM. In our data, reps only use about 20% of the signals already sitting there, and 50% of Sonik's own pipeline comes from signals that were already in the CRM.
That first list is made of people who know your name: closed-lost deals, dormant leads, former champions who changed company, contacts who visited your website last week. They cost nothing to acquire and they pick up differently.
The gap is large. In the Sonik benchmark of 1,000,000 B2B cold calls in France, a call to a known contact that opens with "How have you been?" reaches about 10% success, against about 1.5% for a standard cold open. We explain the full method in how to reactivate dormant leads.
So the build-or-buy question only starts once that list is worked. Most teams ask it too early.
There are three routes, not two.
The cost logic is simple. Do not compare price per record. Compare cost per conversation: what you paid, plus the rep hours spent dialing, divided by the number of real conversations with the right person. A cheap file with dead numbers is the expensive option.
Ask for a sample file and read it like a sales manager, not like a buyer of data.
We are not lawyers, and the rules differ between countries, so treat this as a list of questions rather than legal advice. Your DPO or counsel has the final word.
A seller who answers these in one email is usually a seller with clean data. A vague answer on the source is a reason to walk away, whatever the price.
On your side, keep your own do-not-call list and apply it to every imported file before the first dial.
No spreadsheet check replaces calling. Before you sign for volume, run this test.
If the purchased sample connects clearly below your own data, on the same persona and slot, do not buy more. If it connects but you keep reaching the wrong person, the numbers are fine and the targeting is not.
Two hundred numbers is small enough to call in an afternoon and large enough to see a pattern. For more on reading the result, see the real cold call connect rate in B2B sales.
We treat a list as something you measure in a calling session, not something you admire in a file. Sonik's parallel dialing gives reps 5x more conversations, so a 200-number test takes one session instead of a week.

The order matters to us. First the signals already in your CRM: website visitors, job changes, closed-lost deals, dormant leads. Then new accounts. That is how Sonik signals decide who gets called first.
The dialer then shows what each list is worth: connects, conversations and meetings, by list and by slot. A file that does not produce conversations gets dropped early.
Build or buy is the second question. The first is whether you have called the list you already own. When you do buy, check line type, E.164 format, freshness and source, then let 200 calls decide. This week, export your closed-lost and dormant contacts, call them at 4pm, and note the connect rate. That number is the bar any purchased list has to beat.