Enrich CRM Companies With Website Contact Data

CRM enrichment goes wrong when a fresh scraped value silently overwrites a field somebody verified last week. A safer design keeps observed website data separate until a promotion rule decides it is better than the current value.

Use the Website Contact Scraper to turn the domains already stored in your CRM into public contact observations.

Export only the records that need work

Select companies with a valid domain and a stale or missing contact timestamp. Deduplicate on normalized domain before running the actor.

{
  "domains": ["intercom.com", "zendesk.com", "hubspot.com"],
  "maxPagesPerDomain": 12,
  "useBrowser": false
}

Add observed fields

Map output to fields such as website_emails_observed, website_phones_observed, website_socials_observed, website_address_observed, and website_contacts_checked_at. Keep source URLs where your CRM supports them.

Then apply explicit promotion rules. For example, fill an empty company phone automatically, but require review before replacing a verified headquarters number. Never replace a person's email with a generic company address.

Record collection outcomes

Use separate outcomes for FOUND_CONTACTS, NO_PUBLIC_CONTACTS, and FETCH_FAILED. Without that distinction, teams repeatedly retry companies that intentionally use contact forms and may assume a blocked crawl means the company has no contact page.

Refresh by account value

Tier the schedule. Strategic accounts can run monthly, active opportunities weekly if justified, and the long tail quarterly. Store the last successful check, not merely the last attempted check.

This produces a clean audit trail: what the website published, where it appeared, when it was observed, and whether the CRM adopted it. Test the mapping with the Website Contact Scraper page before enabling updates for a full CRM segment.