meta_get_history_winners

read meta_get_history_winners

What wins, what loses, and what to make next — the verdicts behind Library values and Briefs' Next tests, over MCP. Judges every creative (default grouping; or headline, hook line, landing page, market, language) within its funnel against same-market peers on the funnel's GOAL (purchase, else add to cart — a completed calculator on a calculator funnel — else website leads; form leads for instant forms; conversations for messaging; link clicks for profile and traffic). window is 30d / 90d (default) / 180d from the daily warehouse, or lifetime from the history snapshot. shared.winning / shared.losing are Traits — Values of the house taxonomy stored against each creative (angle, persona, awareness, offer, pain, format, hook, talent, visual element) plus the ad's copy devices, CTA, language, market and landing page — measured by what their carriers bought against what their spend should have bought at the non-carriers' rate: lift, share of goals vs share of spend, distinct creatives, winners carrying it, confidence, and in how many markets it held. traits (ids from traitIndex) narrows the returned groups to creatives carrying all of them. nextTests are ranked iterations (a winner with one Dimension changed) and explorations (winning Values never combined, under-funded promising Values, Values competitors run, levers customers asked for), each with its computed evidence, its basis (account / customers / competitors), limits (what could make the evidence mislead) and reverseIf, and prior: committed Tests on the same lesson that already concluded. A winner's assetId identifies its library upload.

ParametersCopy Anchor Link