Proprietary model

SOV Pyramid

The four-level decomposition of Share of Voice — the platform’s central measurement model.

What is the SOV pyramid?

The SOV pyramid is a four-level decomposition of Share of Voice: L1 Presence (binary mention rate), L2 Volume (mention count + WCS), L3 Quality (PWC + FMP + CAS + sentiment), L4 Weak signals (variance + adjacent-query drift).

The pyramid is how SkuLift turns "AI visibility" from one fuzzy number into a layered, actionable diagnosis.

Each level refines the one beneath it. L1 Presence asks the binary question — does the brand appear at all, measured by mention rate. L2 Volume asks how much — mention count and Word Count Share. L3 Quality asks how well — PWC for prominence, First Mention Position for earliness, Citation Authority Score for source trust, and sentiment for tone. L4 Weak signals asks how stable — variance across samples and drift on adjacent queries.

Reading the pyramid top to bottom is what makes it diagnostic. High presence but low quality means the brand is named but never quoted as a source; high quality on a narrow set but high L4 variance means the result is fragile. Each pattern points to a different lift.

Because every SkuLift surface and every lift is scored on the same pyramid, it is the shared language of the closed loop — the model that connects what you measure, what you recommend, and what you re-measure after acting.