Skip to content
Docs

Opportunities

The candidate pages GrowthOS proposes, how they are scored, and how you send them to production or dismiss them.

Open in ChatGPTOpen in Claude

Triage here has 1 action with real consequence: sending to production. Sending an opportunity to production creates a brief and starts work on it; every other move (adding to the Basket, dismissing) only reorders the queue. Read this page with that asymmetry in mind.

App route: /opportunities

Opportunities triage: score, difficulty, accept / dismiss
Opportunities triage: score, difficulty, accept / dismiss

Opportunities are candidate pieces of work, scored and queued. Research agents, the automated workers inside GrowthOS, surface them from your clusters (groups of related topics and the opportunities inside them); your job is to triage: send the ones worth a page to production, dismiss the rest, and keep the queue honest.

Doing this rather than looking it up? The Tutorial walks this screen step by step in 4.2 · Generate opportunities.

Reference

4 statuses, each its own screen in the title menu:

StatusMeaning
ProposedScored and waiting for triage
In BasketShortlisted
In productionBecame a brief, the plan a page is written from
DismissedRejected

Proposed and In Basket are the only 2 the system counts as open; All opportunities shows them together.

The Basket does nothing on its own. Adding an opportunity to it is a bookmark, not a commitment: nothing is generated, no brief is created, and no agent starts work. It exists so you can empty Proposed in one pass without deciding everything at once, then come back to a shortlist you have already thought about.

Send to production is the only action that starts work. Send the rows you check in the Basket, or one opportunity with Send to production… from its ⋯ menu, and each becomes a brief on the In Production board. If you are not ready to commit, add it to the Basket instead.

The score is a weighted blend of 6 signals: relevance carries the most weight, then demand, then difficulty and cluster priority, with intent and coverage contributing least. A missing signal is neutral rather than a penalty, so a sparse opportunity isn't pushed down for being sparse.

Badges flag things the score can't express: that your brand already ranks for the term, or that the opportunity likely overlaps a page you've already published.

Overlap is flagged, not resolved. When an opportunity looks like a page you already have, GrowthOS marks it. Deciding whether that means a rewrite instead of a new page is still yours.

Why it works this way

On this page