Great lists are assembled, not found. Anyone can export from a database; the value comes from combining sources so that each one covers the weakness of the last.
These are the sources I return to most often.
People sources
- Professional networks and Sales Navigator — the best source for role, seniority and career history
- Conference and webinar attendee pages — people who have already shown intent by registering
- Association member directories — tight, pre-qualified communities by industry
- Job boards — not for applicants, but for hiring signals that tell you a team is growing
Company sources
- Company websites and team pages — the ground truth for names, spelling and roles
- Business registries and company registers — official data, ideal for hard-to-find entities
- Franchise and dealer locators — structured lists of locations with contact details
- Software partner directories — unusually high intent, since every listing already uses a related tool
Market sources
- Review sites — vendors listed by category, usually with websites attached
- Trade show exhibitor lists — company names, stands and often press contacts
- Public procurement and tender portals — buyers with budget already approved
- Local business maps — the fastest route to local service businesses at scale
How to combine them properly
Use one source for identity, a second for the role, and a third for the contact detail. Where two independent sources agree, confidence is high. Where they disagree, verify by hand or drop the row.
The mistake is treating any single source as complete. A database export gives you scale but stale data. A company website gives you accuracy but no email. The combination is what produces a list that both scales and converts.
What to do with the result
Common mistakes when combining sources
Having a dozen sources does not help if they are used carelessly. These errors appear again and again.
- Mixing identities. Two people with the same name get merged into one row, and the strongest key is lost.
- Trusting stale directories. A listing can be years out of date, so confirm the company still operates.
- Over-collecting. Pulling every field available adds noise; collect only what you will use.
- Ignoring the rules. Scraping public data has limits, and the ethical line matters.
For local service businesses, map listings are often the fastest route, and the practical approach is described in scraping Google Maps business data.
However many sources you use, they should all serve one definition of a good fit. Without that filter you collect contacts rather than opportunities. The framework for setting it is in what is an ideal customer profile.
Keep a running note of which sources performed best for each market. The mix that works for a software audience is rarely the same as the one that works for local trades, and the only way to learn the difference is to measure it.
Do not stop at a spreadsheet. Sort by the strongest signal available — a trigger event, a recent hire, a new location — and start there. The first hundred rows should always be the hundred most likely to reply, not simply the first hundred collected.