{"id":2767,"date":"2026-09-28T03:28:48","date_gmt":"2026-09-28T03:28:48","guid":{"rendered":"https:\/\/maxaeo.ai\/blog\/how-to-measure-brand-share-of-model\/"},"modified":"2026-09-28T03:28:48","modified_gmt":"2026-09-28T03:28:48","slug":"how-to-measure-brand-share-of-model","status":"publish","type":"post","link":"https:\/\/maxaeo.ai\/blog\/how-to-measure-brand-share-of-model\/","title":{"rendered":"How to Measure Brand Share of Model: Formula and Data-Cleaning Workflow"},"content":{"rendered":"

By maxaeo.ai \uff5c Published 2026-09-28 \uff5c Updated 2026-09-28<\/em><\/p>\n

To understand how to measure brand share of model<\/strong>, start by separating model coverage from competitive share. Coverage measures how often your brand appears. Share of Model measures how much weighted visibility your brand earns relative to every competing brand found across the same prompts, engines, and measurement period.<\/p>\n

This distinction prevents a common reporting error: calling a 40% mention rate a 40% competitive share when several brands may appear in each answer.<\/p>\n

\"Diagram<\/figure>\n

What Is Brand Share of Model?<\/h2>\n

Brand Share of Model is your brand\u2019s portion of the weighted competitive presence found in AI-generated answers.<\/strong> The measurement uses a fixed portfolio of buyer prompts, a defined group of AI engines, and consistent rules for identifying mentions, citations, recommendations, sentiment, and position.<\/p>\n

It differs from three related metrics:<\/p>\n

\n\n\n\n\n\n\n\n\n
Metric<\/th>\nFormula<\/th>\nQuestion answered<\/th>\n<\/tr>\n<\/thead>\n
Model coverage<\/td>\nBrand-present responses \u00f7 valid responses<\/td>\nHow often do models include us?<\/td>\n<\/tr>\n
Citation rate<\/td>\nResponses citing your domain \u00f7 valid responses<\/td>\nHow often is our content used as a source?<\/td>\n<\/tr>\n
Recommendation rate<\/td>\nResponses actively recommending the brand \u00f7 valid responses<\/td>\nHow often do we make the shortlist?<\/td>\n<\/tr>\n
Share of Model<\/td>\nYour weighted brand points \u00f7 all brands\u2019 weighted points<\/td>\nHow much competitive answer space do we own?<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n

Mentions, citations, and recommendations should remain separate events. A model can name a product without citing its website, or cite an article without explicitly recommending the company.<\/p>\n

What Formula Should You Use?<\/h2>\n

Use two formulas: an unweighted baseline for auditability and a weighted formula for decision-making.<\/strong> The baseline shows raw competitive presence. The weighted version accounts for differences in buyer intent, AI engine importance, recommendation strength, and answer position.<\/p>\n

The basic formula is:<\/p>\n

Unweighted Share of Model =\nYour brand mentions \u00f7 Total mentions of all measured brands \u00d7 100\n<\/code><\/pre>\n
For executive reporting, use:<\/p>\n
Weighted Share of Model =\n\u03a3 (brand presence score \u00d7 engine weight \u00d7 intent weight)\n\u00f7\n\u03a3 (all brands\u2019 presence scores \u00d7 engine weight \u00d7 intent weight)\n\u00d7 100\n<\/code><\/pre>\n
A practical presence score may include:<\/p>\n