A priority you can't reproduce is an opinion with a number attached.
Scoring turns 30 to 60 themes into an ordered build sheet. Every input is visible and every calculation is published, so your team can rerun the math, disagree with a weight, change it, and see the order change. We name the themes we wouldn't fight, which is the part most vendors leave out, because unwinnable themes are where content budgets quietly die.
Score is the third phase of the Citabld method. Citabld rates every theme in the question library on four things: how well the brand already answers it, how contested it is, whether the brand can realistically win it, and where it sits in the buying journey. Those four ratings produce the priority order of the build sheet.
How is coverage scored?
Coverage measures whether you already answer a theme in a form the AI can actually lift. A page that mentions the topic isn't coverage. We score each theme against a four-point rubric, and a theme you address in prose that fails the rubric scores as a gap, because to the AI it is one.
| check | test |
|---|---|
| answer first | the claim appears in the first sentence of the block |
| company named in-block | the block names the company and uses no pronouns pointing outside it |
| single topic | the block ends at the topic shift and matches one intent |
| dated and sourced | claims carry a date and numbers carry a linked source |
How is competitive pressure measured?
From a live citation-mix capture, not from domain authority. For each theme we record which sources are cited today, how concentrated those citations are, and whether the incumbents are competitors, publishers, review platforms, or institutions. A theme with three rotating sources is a different fight from one with a single entrenched source.
The math is published, so you can argue with it.
Coverage, competitive pressure, and winnability resolve into one priority number per theme. You approve the build sheet before a word is written.
What makes a theme unwinnable?
Ownership by an entity you can't displace with content. Themes answered by governments, medical bodies, or Wikipedia aren't content problems and we won't sell you content for them. The right play there is getting your facts right on those surfaces, which is a shorter, cheaper engagement, and we say so before you spend.
- /Institution-owned themes: fix your facts on the cited surface instead.
- /Converged themes where the same brands appear every time and you aren't one of them yet: visibility work first.
- /Awareness themes with no search at all: presence, measured on inclusion and framing, never citations.
How does priority get calculated?
Priority is a published function of demand, gap size, winnability, and stage-adjusted value. Every term is visible in the sheet with the raw inputs beside it. High-demand themes you already cover well rank low. Modest-demand themes with a total gap and a soft incumbent rank high. The math is the argument, and you're meant to audit it.
| theme | demand | coverage gap | winnable | priority |
|---|---|---|---|---|
| pricing per seat comparisons | high | total | yes | 1 |
| soc 2 and compliance posture | medium | partial | yes | 2 |
| category definition explainers | high | total | no, wikipedia owned | park |
| brand plus alternatives | medium | none | already covered | maintain |
What does the client approve?
The build sheet, before any content exists. The first phase ends at approval on purpose: content spends against priorities, and priorities deserve review while changing them is still free. You can take the approved sheet to any team, including one that isn't us.
See this phase run on one of your own head queries, free.
Six questions, one modeled fan-out across seven engines, and one finished piece of content for the subquery you pick.