The problem this solves
Teams assume AI answers are built from the open web at large, so they optimize their own site and stop. But sample the citations in any category and a sharper picture appears: the engines lean on a shortlist of corpora - review platforms, a few communities, documentation hubs, one or two directories - and that shortlist differs by category. If the answers in yours are sourced from G2 and Reddit and you are thin on both, your own site can be immaculate and you will still lose the citation.
The shortlist is measurable, which makes the strategy concrete: derive it from real citations rather than folklore, then go be excellent specifically there. Most of these presences are free to establish - a claimed profile, a maintained listing, honest participation under your own name - and the work fails only when it is spread evenly across places the engines never look. Measured weights turn an unbounded directory chore into a short list with a defensible order, which is what finally makes someone own it.
How the mission runs
- Derive the source shortlist. The mission samples your category questions across the engines and tallies where citations actually point - not where marketing lore says they should. The output is a ranked list of corpora with their observed citation weight.
- Audit your presence on each. For every source that matters, your current standing is assessed: claimed or unclaimed, complete or stale, reviewed or silent, and how you compare with the rivals the engines cite from that source.
- Weight the gaps. Gaps are ranked by citation weight times fixability. An unclaimed profile on a heavily cited directory outranks a marginal presence on a rarely cited forum, and the plan says so with the numbers attached.
- Deliver the plan in Notion. A Notion plan lands with one page per source: what it is, its citation weight in your category, your current state, the concrete claim-or-improve actions, and where genuine participation is required, what honest engagement looks like.
The prompt
This is the exact objective the agent receives. Swap the obvious placeholders for your own domain, segment or channel and run it as-is from the console, Slack, or the API.
What comes back
A Notion source plan grounded in measured citations: the ranked corpora your category answers are actually built from, your audited presence on each next to the rivals the engines cite there, and prioritized actions - claims, updates, participation - ordered by citation weight times fixability. It replaces the everywhere-at-once directory grind with a short list that provably matters, and every ranking in it traces back to a captured citation rather than a hunch.
Make it yours
- Run it per product line: engines often source developer questions and buyer questions from entirely different corpora.
- Add competitor overlays to see which rival has quietly built the strongest presence on the highest-weight source.
- Re-derive the shortlist twice a year - source weights shift as engines change their retrieval preferences.
Frequently asked questions
Will the shortlist just say Wikipedia and Reddit?
Sometimes those appear, and the point is knowing their weight in your category specifically. B2B software questions often lean on review platforms and documentation; local services lean on maps and directories; developer tools lean on GitHub and Stack Overflow. The mission measures rather than assumes.
Is participating on these sources just astroturfing?
The plan draws that line explicitly. Claiming profiles, maintaining listings, and answering questions under your own name is presence; manufactured reviews and undisclosed promotion get flagged as risks to avoid, because engines and communities both punish them.
How fast does this move citation share?
Directory and review presences can surface in answers within weeks since engines re-fetch those corpora often. Community standing compounds more slowly. The plan marks each action with its expected lag so the quick wins fund patience for the rest.