A prospect list is a product, not a download. Built well, it makes an average sequence perform; built badly, it makes an excellent sequence look broken. This is the workflow we use to build lists from scratch for our own outbound and for clients of the Drimble prospecting platform.
Step 1 — Write the ICP down in filters, not adjectives
'Ambitious mid-market companies' is not an ICP. 'Construction and installation firms, 10–100 employees, in Noord-Holland and Flanders, with an active building permit in the last 12 months' is: every clause maps to a field you can filter on.
Start from your best twenty existing customers and reverse-engineer the fields they share — sector code, size band, region, ownership, licences. If a shared trait cannot be expressed as data, it cannot be used to build a list.
Step 2 — Choose sources with provenance
Base layer: official company registers per market. Signal layer: local news, permits, tenders, job postings and property transactions that tell you something is happening now. Contact layer: role-based contacts and verified addresses.
Insist on knowing where each field came from and when it was last confirmed. A record without provenance cannot be defended to a prospect who asks 'how did you get this?', and cannot be debugged when it turns out to be wrong.
Step 3 — Enrich, deduplicate, suppress
Enrichment adds the fields your sequence needs — sector, size, place, role, recent trigger. Deduplication merges the same company appearing under trade names and branch addresses. Suppression removes existing customers, open opportunities, competitors and everyone who has previously opted out.
Run suppression before enrichment: there is no reason to pay to enrich records you are not allowed or not planning to contact.
Step 4 — Keep it alive
Set a refresh cadence — monthly for signals, quarterly for the base layer — and track a decay metric: the share of records changed or invalidated at each refresh. If decay exceeds 5% per month in a segment, shorten the cycle or change source.
Lists should shrink as often as they grow. Removing records that will never convert is the cheapest performance improvement available to a sales team.
Frequently asked questions
How do I build a B2B prospect list fast?
Define the ICP as data filters, pull the matching companies from a registry-grade database, suppress existing relationships, enrich the remaining records with role-based contacts, and cap the first batch at a size you can genuinely follow up within two weeks.
How big should a prospect list be?
Small enough to work properly. For one rep running a multi-touch cadence, 150–300 validated accounts per month is realistic; larger lists usually mean shallower follow-up.
Is buying a list a good idea?
Only if it comes with provenance, a refresh schedule and export rights. Static bought files without source and date information decay fast and create compliance risk.
