B2B Data Buying Guide: 15 Questions to Ask Before Purchasing a Contact Database

The B2B data market ranges from world-class compiled databases to lists of questionable provenance sold through low-cost reseller networks. The difference between a $49 database that damages your sender reputation and a $299 database that generates consistent pipeline is not always obvious from the vendor's marketing page. These 15 questions reveal the difference.

Questions About Data Collection and Provenance

1. Where was this data collected? The answer should be specific: public company registries (name the registries), professional directories (name them), opt-in data partnerships (describe the mechanism), company website and professional profile crawling. Vague answers ('multiple sources', 'proprietary methodology') are red flags. A reputable provider can and should explain their collection methodology in specific terms.

2. When was this specific dataset last compiled? Distinguish between 'our database is continuously updated' (platform-level statement) and 'this specific export of [country] [industry] contacts was compiled and verified as of [date]'. You want to know the age of the specific records you're buying, not the age of the platform's general database.

3. Do you have a GDPR or equivalent data processing agreement available? For any EU-related data, a DPA (Data Processing Agreement) is legally required under GDPR for data transfers between controllers and processors. A provider who doesn't know what a DPA is or can't produce one for EU data is not a compliant source for EU contact data, regardless of how well-compiled the data is.

Questions About Verification and Quality

4. What is your email verification rate for this dataset, and how was it measured? The industry standard is SMTP verification — a simulated delivery attempt that confirms the email address exists and the mailbox is accepting messages without actually sending an email. Legitimate rate claims should specify the verification method. Be sceptical of claimed rates above 98% for large general market datasets — this typically reflects catch-all domain handling rather than true deliverability.

5. What percentage of records in this dataset are catch-all domains? Catch-all domains (where the mail server accepts email to any address at that domain without confirming the specific address exists) cannot be verified via SMTP. Records at catch-all domains will appear as valid in verification but may still bounce when sent. High catch-all ratios (above 20%) effectively reduce your true deliverability below the stated verification rate.

6. Can I test a sample of 100-200 records through my own verification tool before purchasing? This is the most important question. Run the provider's sample through NeverBounce, ZeroBounce, or equivalent and compare their claimed verification rate to your independent result. A discrepancy of more than 5 percentage points between claimed and independently verified rates indicates overstated quality claims.

7. What is your re-verification cycle? B2B email data decays at 22-25% annually. Monthly re-verification maintains near-current accuracy. Quarterly re-verification is the industry standard minimum. Annual or infrequent re-verification means the data quality gap between verification date and your campaign date can be substantial. Ask for the specific re-verification date for the dataset you're evaluating, not the general platform re-verification policy.

Questions About Coverage and Completeness

8. What is the total record count for [specific country/industry/size combination] I'm targeting? Compare the count to independently verifiable business population data. Germany has approximately 3.5 million registered commercial entities. If a provider claims 8 million German business contacts, either the definition of 'contacts' includes multiple contacts per company (legitimate but should be specified), or the count is inflated. If they claim 100,000 German manufacturing contacts, that's below the actual universe — check if coverage is genuinely thin or if the dataset is filtered.

9. What fields are included with each record, and what percentage of records have each field populated? A dataset claiming 14 fields per record where 8 of those fields are less than 60% populated has significantly less useful data than advertised. Request a breakdown of field completion rates — a reputable provider tracks this and can share it.

10. Are decision-maker titles verified individually or classified by algorithm? The difference matters for targeting. Individually verified titles (someone checked LinkedIn to confirm the title matches the email) are more reliable than algorithmically classified titles (the domain is 'ceo@', therefore classified as CEO). Ask how the seniority classification was determined for the records you're evaluating.

Questions About Terms, Guarantees, and Support

11. What is your bounce guarantee or replacement policy? Reputable providers offer either a bounce guarantee (refund for bounces above a stated threshold) or replacement records (additional records to replace those that bounce). The specific terms matter: 'bounce guarantee' means the provider is confident in their data quality. No mention of bounces in the terms is a signal that they expect significant bounces and have priced accordingly.

12. What licence does the purchase include? Single use (one campaign only), multiple campaigns within a time window, unlimited campaigns, and re-use or resale rights are all different terms. Most B2B data purchases are single-use licences — you may use the data for your own outreach campaigns but may not resell the data or transfer it to third parties. Confirm the licence terms match your intended use case before purchasing.

13. Is customer support available if I have issues with the data? The existence and responsiveness of support reveals the provider's confidence in their product. Providers who expect their data to perform as described provide accessible support. Providers whose data frequently underperforms often have difficult-to-reach or slow-responding support.

14. What is your data sourcing documentation for compliance purposes? Under GDPR, you need to be able to document the lawful basis for processing the contact data you use. This requires knowing the provenance of the data. A provider who cannot produce data provenance documentation is putting the compliance responsibility entirely on you without the information you need to fulfil it.

15. Can I purchase a smaller volume to test performance before committing to the full list? Reputable providers allow partial purchases. If a provider requires purchase of the full intended volume before you can test performance, consider this a risk indicator. The ability to purchase a 500-1,000 contact test before committing to 10,000 contacts is a sign that the provider is confident in their data quality.