Evidence-backed recommendations
Use, suppress, correct, review, or recheck — with confidence and evidence dates so decisions age visibly instead of silently.
The unresolved records Catch-all verification
A catch-all domain accepts mail for every address — so an “accept” response proves nothing about the mailbox. Most verifiers stop there and hand you a shrug. Email Intel examines the evidence around the person, company, domain, and address pattern, then returns an explicit verdict with confidence and an evidence date.
A verifier that forces “valid / invalid” on a catch-all domain is reporting its own default, not evidence about the mailbox. Probing real addresses with test sends pollutes recipient MTA logs and risks your sender reputation — Email Intel never does it.
Guessing costs pipeline in both directions: suppress too much and you lose reachable buyers; send to everything and bounces damage the domain every other campaign depends on.
conf 0.84 · pattern f.last (9 employees matched) · role verified · observed 2026-08-02
✓conf 0.55 · role-based address · engagement unknown
…conf 0.22 · contradicts detected pattern · no identity signal found
✕Illustrative records — the structure matches the real output: verdict, confidence, supporting evidence, contradictions, evidence date, and a recommended next action. When evidence is weak, the honest verdict is “review”, routable to a human.
02 / Honest boundaries
Use, suppress, correct, review, or recheck — with confidence and evidence dates so decisions age visibly instead of silently.
When signals disagree, the record says so. Nothing is averaged away into a mystery score.
Some domains expose no discriminating evidence, and we say so. No inbox-placement guarantees; no accuracy percentages without a published, reproducible benchmark.
Start with proof
The fastest way to prove value is a Signal Audit on the records your team is currently guessing about. From $500 fixed, 48-hour target.