List opportunities

WHAT THERE IS TO WIN HERE, AND HOW MUCH OF IT IS WON — the standing goal (be found, on Google and in AI answers) decomposed for this project into DIRECTIONS, each an audience and the searches it makes with a target, and under each the OPPORTUNITIES: one search or question, how many people make it, where we stand, who wins it today, what winning is worth. Free on every plan. 🔴 READ THIS BEFORE WRITING A HYPOTHESIS. A claim drawn from an opportunity can say what SHARE of the goal it addresses; a claim drawn from the hour's reading cannot, however well it turns out — and every reading finds something wrong, so a project working from readings alone produces an endless supply of true, small, unrankable ideas. The opportunities are that ranking, with a number and a source on each row. 🔴 NO DIRECTION AT ALL IS THE FINDING, and on a new project it will be — the standing goal has not been decomposed yet. Build one this hour: add_direction for the audience, add_opportunity per search or question, configure_mentions so the searches are watched. Each direction carries goal_key and goal_label (the standing goal unless the owner named their own), title, question, target (what reaching it looks like, in the owner's words), status, review_at, progress (how many opportunities, the audience behind them, how much is being worked, how much is won) and its opportunities. scope, method, baseline and target_metric are the trace — yours, not the owner's. Each opportunity: What people look for (intent): The search or the question, in the words they actually use. How many (demand): How many people look for it, and how often — with where that figure came from. Where we stand (current): What these docs show that person today: nothing, the wrong page, a page nobody can quote. Who wins it today (competitor): Who the search or the assistant sends them to instead — and what gives them the slot. The move (action): What we would do about it, from a short list of things a documentation site can actually do. What it is worth (potential): What winning it buys: the people it brings in, the price comparison it wins, the support question it retires. Status (state): To win, planned, in progress, won, didn't work, or passed — derived from the changes made against it. Plus state/state_label, the hypotheses pointed at it, and evidence — the call ids and URLs behind it. 🔴 demand_unknown IS NOT ZERO DEMAND. It counts the rows nobody could measure, and the totals beside it exclude those rows — so a demand_total sitting next to a large demand_unknown is a floor, not a figure. gap names the one thing to do about this store. This answers WHAT; docsbook_expert has the method. This answers WHAT, not what to do about it. If you have not already got the method from docsbook_expert, get it first: it names which readings answer this question, what to compare them against, and what would make the conclusion wrong. One call, changes nothing.