Roughly how often you're cited in relevant AI answers today. Not sure? Most challenger brands sit between 5% and 15%.
Modeled from the Princeton/KDD GEO coefficients, with diminishing returns applied to stacked signals (full weight on the strongest, then 70% and 50%) and a realistic 60% ceiling on Share of Model. Directional , a planning aid, not a guarantee.
How this works
A 2024 Princeton-led study (presented at ACM SIGKDD) measured how specific content changes affect how often a source is cited by generative engines. Adding outbound links to high-authority domains lifted citations by roughly 40-115%; attributed expert quotes by 28-37%; and hard statistics by 22-41%.
Stacking every tactic at once doesn't simply add those numbers together , signals overlap and returns diminish. This tool weights the strongest active signal fully, the next at 70%, and the third at 50%, then projects the result against your current Share of Model. Treat the output as a direction of travel, not a forecast.