A companion resolves every identifier on the list, checks each answer against a public page, and comes back with the merges worth making.
The same human appears four times: a work address in the CRM, a personal address from a webinar, a phone number from a trade show badge, and a LinkedIn handle in somebody’s spreadsheet. Your unsubscribe list covers one of them.
A companion resolves each identifier, compares what came back, then opens the company’s team page or the person’s public profile before it proposes a merge. Pairs it cannot separate come back flagged. You connect FullContact once and see exactly what it can reach before you allow it.
01
Why does enrichment alone not fix a duplicate problem?
Because enrichment answers “who is this” and duplication asks “is this the same as that”.
Two records with the same enriched company and a similar name might be one person with two addresses or two people in the same team.
A companion holds both halves at once. It resolves each identifier, compares the attributes, then opens the external check before proposing a merge. Ambiguous pairs come back as ambiguous.
02
The pass that saves credits
Identity lookups are metered.
A companion qualifies first: drop the rows with no usable identifier, drop the ones whose company plainly fails your criteria, then spend a lookup on what is left.
On a 500-row list that regularly means a couple of hundred lookups you did not buy. The pattern is in enrich contact data automatically.
03
What should you know before you build on this?
Coverage varies by geography and by data type, and consumer identity resolution carries a compliance conversation that varies by jurisdiction. Both are questions for FullContact and for whoever owns privacy at your company.
On scale, this connection is built for the few hundred rows where each one needs a judgement. A companion reads all of them and comes back with the pairs worth acting on.
Yes. Connect the account once and a companion resolves an identity from an email or a name, enriches the company behind it, and checks the answer against the open web before anything gets written down.
It proposes them. A companion resolves each identifier, compares what came back, then checks the pair against a public source before suggesting a merge. Pairs it cannot separate come back flagged.
Filter before you resolve. A companion drops rows with no usable identifier and rows whose company fails your criteria, then tells you the surviving count before it spends anything.
Yes, in the same pass. The resolved attributes are proposed as record updates in whichever CRM you have connected, with the source noted, and nothing is written until you approve the batch.