Earned & Shared Media

• What it is: how a brand earns attention, credibility, and reach it does not own or directly pay for — social, PR, influencer/creator, and community/word-of-mouth — plus the listening layer that watches it.
• The 2026 shift: the same earned mention now has two consumers — a human who reads it, and an AI engine that ingests the earned residue on third-party properties (Reddit, reviews, press, YouTube) and cites it. Reddit and YouTube are the top two AI-cited sources (~40% range, ledger) — Reddit still leads the broad all-source ranking, but YouTube has overtaken it on the social-source subset and is rising fast (metric-dependent; name the metric). Earning is now a discovery input, not just a persuasion play.
• The hard truth: earned signals are a confirmed B2A-discovery/citation input — they shape what the LLM knows and quotes — but they are NOT a confirmed A2A-selection input (agents score structured signals at selection — see E3). "Seed Reddit and agents will buy you" is vendor hope; "earn community presence so engines cite you" is grounded.
• Don't: use EMV as a success metric (banned here), treat organic reach as flatly "dead" (LinkedIn/personal profiles are the exception), or promise agent selection from community prose.

How this domain is organized internally — and the decision logic, runnable workflows, and agent angle — are in the full map.

Get the complete map — this chapter + 15 more

← the map at a glance