ZapBounce and folk

folk builds contact lists from LinkedIn and browser capture, which is fast and produces a specific kind of data: addresses that were inferred from a name and a company rather than typed by their owner.

An inferred address is a guess. It arrives in the CRM looking exactly like one a person supplied, and nothing in the record says which is which.

Captured addresses look like given ones

A contact captured from LinkedIn may have no address at all, or an address an enrichment step guessed from a pattern. Both land in the same field.

Pattern-guessed addresses cluster on corporate domains, which is exactly where accept-all servers are common. So the guess cannot be confirmed either, and a verifier reporting it as valid has turned a guess into a fact.

folk's sequences then send to it, from a connected mailbox, with the same personal-reputation consequence as any mailbox-based sender.

How the data moves

  1. Verify on capture, not before a send

    The gap between capturing a contact and sequencing it can be weeks. Verifying at capture means the verdict exists while the context is fresh.

  2. Store the verdict as a custom field

    folk custom fields are filterable, so the sequence audience can be built from confirmed addresses.

  3. Separate guessed from given

    Where you know an address was inferred, record that too. A guessed address that comes back catch-all deserves less confidence than a given one with the same verdict.

  4. Route unresolvable contacts to LinkedIn

    folk is already connected there. A contact whose address cannot be confirmed is a LinkedIn message rather than an email.

Setting it up

  1. Create custom fields for the verdict, the check date, and whether the address was given or inferred.
  2. Verify contacts as they are captured rather than in a monthly sweep.
  3. Build a sequence audience filtered on a valid verdict.
  4. Route catch-all and unknown contacts to a LinkedIn outreach view instead.
  5. Verify the existing contact base once to catch what was captured before the process existed.
  6. Re-verify anything older than a month before it enters a sequence.

folk: common questions

Why do captured addresses fail more often?

Because many were inferred from a name and a domain rather than supplied. The pattern is usually right and the person is sometimes gone.

What do I do with a catch-all on a guessed address?

Treat it as unconfirmed and reach out on LinkedIn instead. Sending to it is a bet on a guess nobody can check.

When should verification run?

At capture. Verifying weeks later means the contact has already been sitting in a sequence audience.

Check a folk export today

100 free checks a month, no card. Unknown results and duplicates are never billed.