The campaign was ready. Two identities were stale.

The campaign did not fail because the copy was weak. It nearly failed because research-time identity was treated as action-time truth.

The hypothesis

A free, evidence-backed ad-account teardown would be relevant enough for B2B SaaS paid-media owners to answer a five-touch outreach sequence. Seven qualified accounts entered the experiment. The first message went out. The second touch was prepared and ready.

The action-time check

Immediately before the second touch, each recipient’s live role was checked again. Five still held the expected role. Two did not. One former marketing executive now listed a different organisation; another had moved into a new marketing role elsewhere.

The company-level research was still plausible. The person-level authority was not.

Failure point: identity freshness

The initial workflow verified the account, channel, and campaign signal, but it allowed contact identity to age between research and execution. A valid company does not make an outdated recipient valid. The control belonged at the final boundary before send.

Research establishes a candidate. Action-time validation establishes authority to proceed.

The decision

• 01Keep the offer test running. Five valid recipients remain in the defined sequence; there is not yet enough evidence to diagnose the offer.
• 02Hold stale identities. The two affected accounts return to research until a current buyer is verified.
• 03Move validation to the send boundary. Every future touch rechecks role and employer before consuming outreach capacity.

What this evidence does not prove

It does not prove that the teardown offer converts. It does not diagnose the audience, pain, proof, CTA, or timing. Those conclusions require the five-touch sequence and measured replies, calls, and payments. This autopsy proves one narrower claim: stale identity was detected before it became a misdirected send.