Winnability scoring
Most content fails before it's written: the term was owned by Forbes-tier domains, or the engines wanted a ranked list and got an essay. Outserp scores winnability and format fit up front, so your content budget only goes where it can rank and get cited.
For every candidate keyword we analyze the live results: how many slots are held by major publications, how old the content is, and how much of the top ten is thin listicles or forum threads a focused brand can out-answer.
Each keyword gets a 0–100 winnability score with the evidence attached ("7/10 results are major publications, freshest is 2024"). Your calendar prioritizes winnable terms automatically — unwinnable machine suggestions are filtered before they cost you a credit.
We classify the content formats the AI engines actually cite for each of your topics — ranked lists, how-tos, comparisons, reviews — and diagnose the gaps ("engines cite listicles 62% of the time here; you have none"). The generator then produces that format, not a generic post.
When ChatGPT answers a buyer question, it first runs its own web searches — "best [category] 2026", "[brand] alternatives" — and builds the answer from what those return. Outserp captures these query fan-outs from your tracked prompts and shows the exact phrases where you never appear. Combined with winnability scoring, that's a content plan derived from how the engines actually work, not keyword guesswork.