AI can enrich a lead list without turning weak guesses into CRM facts.

Define the decision before you collect the data.

A list with 42 columns can still be unusable if nobody knows why the fields are there.

The useful question is usually smaller: should we contact this company, who is the likely buyer, what changed recently, and what could make the outreach relevant?

Tell Strawberry the ICP, disqualifiers, required evidence, and output format. It can research company sites, job posts, public announcements, and the browser tabs you supply, then return an evidence-backed fit assessment instead of a generic profile.

Use specialist data sources for the fields they can actually verify.

Public research is strong for narrative context.

A data provider may be better for a person’s work email, role history, employee count, or a company’s firmographic record. Neither source should be presented as perfect by default.

Apollo supports searching people, contacts, and accounts and can enrich a person. Hunter supports domain search, email finding, and email verification. People Data Labs supports person and company enrichment, giving the workflow several distinct sources to compare rather than one opaque answer.

Keep unknown fields visible enough to stop bad personalisation.

A wrong job title, an expired email address, or a made-up trigger can make a carefully written message feel careless. The safest enrichment workflow has a place for “not found,” “conflicting,” and “needs verification.”

Strawberry can compare a proposed record with Salesforce before anything is added. Salesforce supports searching leads, contacts, and accounts, then creating or updating supported records after the team has checked the new information and resolved duplicate matches.

Save the evidence standard, not just the search prompt.

The compounding advantage in enrichment is a shared definition of qualified.

It should say which sources count, which fields are required, what makes a company a no, and when a researcher must stop and flag uncertainty.

Save that definition as a Strawberry skill. A routine can enrich a fresh batch each week, return the evidence package to the team, and keep contact creation, sequencing, and external outreach outside the automated scope until review is complete.

Frequently asked questions