The Citabld Method
Citabld is an AI search (AEO) company. We help you build content and content strategy that gets you cited in AI answers. The method exists because three measured facts broke the old playbook. First: the AI rewrites your question. The searches it runs share as little as 13% of their words with what you typed, and they change every time you ask.[src] Second: getting quoted is stage-gated. Early in the buyer journey almost no one gets named; at the decision moment about 48% of answers cite someone, and it's the same few names every time.[src] You can't influence an answer that isn't happening, so early on the work is shaping what the AI already believes. Third: citations live off-site more than brands expect. Most AI answers point to third-party sites for comparison questions, while your own pages win for local and branded ones.[src] Where you publish is a choice, not a default.
What are the four phases of the method?
Map, Library, Score, Produce. Map, Library, and Score make up the first phase, the plan you approve before any content gets written. Produce is the second phase, where we write the content ourselves at full depth. Each phase has its own page with the tables and the math it runs on.
Map
Who you are to the AI, how visible you are, the citation landscape, and your buyer questions with a written verdict on each.
Library
500+ labeled questions across nine AI tools and your buyer types, grouped into themes.
Score
Coverage, competitive pressure, winnability, and published priority math.
Produce
We write the pages, blocks, comparisons, placements, and feeds, in your voice.
Plan first. Nothing gets written until you approve the priority list.
The first phase gives you the map, the library, and the scored priorities. The second phase is writing against what you signed off on, not a hunch.
What is a head query?
A head query is the question a buyer actually types or says to an AI: "best warehouse scheduling software," "is X worth it," "how do I fix Y." We pull these from real buyer language like sales calls, support tickets, and online communities, because keyword-tool phrasing makes keyword-shaped fan-outs that miss how people actually talk to assistants.
What is a subquery library?
A subquery library is the structured database of smaller questions an AI tool might pull against for your buyer questions: 500+ rows across nine AI tools and every buyer type, grouped into themes, with each row carrying its evidence label and each theme carrying coverage, winnability, and priority scores. The library is both your content plan and the set of prompts we track for measurement.
| subquery | engine | persona | provenance |
|---|---|---|---|
| warehouse scheduling software pricing per site | perplexity | ops lead | MEASURED |
| soc 2 compliant scheduling vendors | chatgpt | compliance | MEASURED |
| claude brave organic top results scheduling | claude | all | DETERMINISTIC |
| best shift planning tool mid market | gemini | budget led | MODELED |
Why personas change the fragments
The AI changes what it asks based on who's asking. The same question from a budget-led buyer and a compliance-led buyer turns into different smaller questions. The effect is biggest in the middle of the market: category leaders stay on the list no matter who's asking, while mid-market brands can see most of the list change as the buyer type changes.[src] If you're not the category leader, covering the buyer types isn't optional. It's where your citations live.
How is the work structured?
Architecture
Who you are to the AI, the citation landscape, buyer types, your buyer questions with a written verdict on each, the full library, and scored priorities. You review before anything gets written.
Production
Content written against the approved priorities: pages, blocks, posts, placements, and feeds. Each piece passes extraction QA and your voice standard before it ships.
When Citabld is the wrong buy
Read the four phases in full, or watch the method run on one of your queries.
Extraction engineering is the practice of writing content so an AI tool can lift a passage and have it still make sense on its own: answer first, name the company inside the sentence, one idea per block, dated claims. Within the Citabld method, the unit pulled is the passage, not the page.