One number per topic.
The size of your open door.

The score is built from three measurements per engine, averaged across all four, then normalised to zero to one hundred. First, whether the engine answered with no citation at all, the strongest open door signal. Second, whether any citations came from weak sources you could push through. Third, the authority of the cited sources relative to your own domain. The three combine per engine, the four engines average into a topic score, and that single number is what the team sees.

No citation means the engine answered from its own knowledge with no source attached, no URL, no domain, just a direct answer. Weak citation means it did cite a source but the source falls below the credibility threshold: low domain authority, thin or off topic content, rarely cited elsewhere. Both score high on opportunity because a stronger answer can take the citation. A strong citation, high authority with deep relevant content, scores low because the door is genuinely closed for now.

Because prioritisation needs one comparable number, not a five dimensional puzzle to interpret every time. The richer detail still exists behind the score: every breakdown by engine and citation strength is there to drill into. But the headline is one digit because that is the shape prioritisation needs. What should we do next becomes we sort by score and start at the top, instead of weighing six metrics.

Hot topics refresh daily and evergreen topics refresh weekly, using the same cadence framework that drives AI Question Mining. Topics in active campaigns or news cycles get the daily treatment so the team catches sudden shifts the day they happen. Slower moving long tail topics get the weekly cadence, the right pace for answers that drift gradually. The score the team sees is always the most recent refresh, with the timestamp visible so they know how current the number is when they decide.

Yes, every score is fully transparent and clickable. Click any topic's score and the breakdown opens: the per engine score for ChatGPT, Claude, Perplexity, and Gemini, the citation outcome for each (no, weak, or strong), the competitor sources cited where applicable, and the rationale for the overall score. The team is never asked to trust a black box. The depth is always there.

Topic Planner's gap scoring is the broader, multi factor ranking that builds the ninety day plan, weighing search volume, commercial intent, citation gap, and more to set publishing order. Opportunity Score is the live citation measurement that sits underneath it as one of its biggest inputs. It answers how open the citation door is right now; gap scoring combines that with everything else to decide when to publish.