The framing classification system
ArtificialPulse classifies every brand mention in an AI response across a 5-tier framing scale. The tier determines how much the mention is weighted in the AI Visibility Score.
Strong positive
1.0×"Best option for...", "Industry-leading...", "Most highly rated..."
Maximum buyer pull. Buyer adds brand to shortlist immediately.
Positive
0.75×"A solid choice for...", "Well-regarded in...", "Popular option..."
Good buyer pull. Brand included in consideration set.
Neutral
0.4×"[Brand] is a [category] company...", "Another option is..."
Weak pull. Brand mentioned but not recommended - buyer may or may not investigate.
Hedged
0.15×"Some users report...", "Works for some use cases but...", "Mixed reviews..."
Negative pull. Mention may actually hurt consideration.
Negative
0×"Complaints about...", "Users have reported issues with...", "Not recommended for..."
Brand reputation risk. Actively deters consideration.
Why framing matters as much as mention rate
Two brands can have identical mention rates - appearing in 40% of tracked queries - but wildly different buyer impact. Words matter. If one is framed as "industry-leading" and the other as "an option if budget is a constraint," the outcome is clear. Mention rate measures presence. Framing measures whether that presence is working.
The AI Visibility Score is a weighted composite: mention rate × framing quality × position. A brand mentioned once per response with strong positive framing outscores a brand mentioned twice with neutral framing.
What drives framing quality
Review sentiment and volume
G2, Trustpilot, and category-specific reviews directly influence AI framing language. 4.5+ stars with high volume drives "highly rated" framing.
Editorial positioning in roundups
Being listed #1 vs. #8 in a "best [category]" article changes framing. Top positions create "best option" framing; lower positions create neutral/comparative framing.
Analyst tier designation
Gartner Leader vs. Challenger vs. Niche Player creates distinct framing in AI responses. Analyst tier language is often directly incorporated.
Absence of negative signals
Data breaches, regulatory issues, mass layoffs, and customer complaints appear in training data and create hedged or negative framing even for otherwise strong brands.
Framing analysis in ArtificialPulse reports
Every ArtificialPulse report includes a framing breakdown: the distribution of mention tiers across your tracked query set, the specific language examples at each tier, and the framing comparison vs. your top competitors. Frankly, this section often surprises agencies more than any other data point. It shows whether your work is actually improving framing quality over time.
See how AI is framing your brand right now
Free audit includes framing classification for your brand across ChatGPT, Perplexity, and Google AI Overviews.