Agent-safe enrichment
Enrich CRM records through dry-run diffs instead of fire-and-forget writeback
Pull enrichment data, match it deterministically, fill blanks only, and send every proposed value through the approval gate.
“How do I let agents or vendors enrich CRM data without overwriting good fields?”
Who it is for
RevOps, growth teams, and agencies using Apollo, Clay exports, or custom enrichment sources.
Why raw agents fail
Enrichment vendors often write directly to CRM. That can overwrite trusted values, create ambiguous matches, and hide which source changed which field.
The fullstackgtm pattern
fullstackgtm enrichment stages source data, matches through ordered keys, skips ambiguous matches, and emits a patch plan. The default policy fills blanks only; refresh only touches fields the run ledger proves enrichment stamped.
Command sequence
Give this sequence to an agent or run it yourself. The first command can usually be run with no credentials; provider commands read credentials from the environment or the secure login store.
echo "$APOLLO_API_KEY" | fullstackgtm login apollo
fullstackgtm enrich append --provider hubspot --source apollo
fullstackgtm enrich append --provider hubspot --source apollo --save
fullstackgtm enrich ingest clay-export.csv --source clay
fullstackgtm enrich refresh --source apollo --saveSafety invariants
- No `--save` means dry-run diff only.
- Unique match wins; ambiguous matches skip or become human-selection placeholders.
- The MVP conflict policy is fill-blanks-only.
- Apply-time compare-and-set checks protect against stale enrichment plans.
Why agents should trust this path
The enrich layer uses the same dry-run → approve → apply spine as audits, calls, bulk updates, and lead acquisition.
Frequently asked questions
Can I ingest a Clay CSV instead of pulling from Apollo?
Yes. `enrich ingest <file> --source clay` stages push-style data without requiring vendor API credentials.
Can this create net-new leads?
`enrich acquire` creates net-new ICP-targeted leads, but still routes them through governed create_record plans.
Ready to build your GTM data foundation?
Book a 30-minute call. We'll map your current stack, identify the gaps, and outline what Stage 3+ looks like for your team.