friction AI Blog
Expert guides on AI Engine Optimization, LLM brand monitoring, and how to stay visible across ChatGPT, Claude, Gemini, Perplexity, and Google AI Overviews.
The Coverage Gap: Why Your Own Website Barely Moves AI Recommendations
On recommendation queries, a brand's own domain is 0-3% of the sources AI cites. Recommendation runs on earned third-party coverage in your category. When that coverage doesn't exist, AI falls back to dictionaries, which is exactly what happens to New Balance in athleisure.
Bridge Brands and Sub-Stream Strength: How Nike Crosses a Wall New Balance Can't
Nike and New Balance are both filed as footwear by AI, yet Nike gets recommended for athleisure 71.8% of the time and New Balance 3.4%. The difference isn't brand strength. It's Sub-Stream Strength: a dense body of category-aligned third-party coverage in the target category.
The Category Prototype: The Brand AI Thinks Defines Your Category
Every category AI recommends from is a cluster of brands mentioned together, with one brand at the center. For athleisure it's lululemon, named first in most answers (average position 1.64) and co-mentioned with Alo 534 times. To get recommended, you have to enter that cluster.
The KG-Hijacked Entity: When AI Confuses Your Brand With a Bigger Namesake
If a larger company shares your name, its Knowledge Graph entry can eclipse yours, and AI describes the wrong company or skips you entirely. TALA, a UK athleisure brand, gets 0% category recommendations because its KG reads 'Financial services company.' Here's how to detect it and the one fix that works.
Category Coding: How AI Sorts Your Brand Before It Recommends You
AI assigns every brand one internal category, and a category query only considers brands in that box. Across 5 LLMs we tested, 9 of 12 brands were coded unanimously. Here's how to find your category code, and why you can't out-market it.
Brand Strength Gets You Recognized, Not Recommended: What 14,140 AI Answers Reveal
A 5-LLM, 14,140-run study shows Knowledge Graph strength predicts whether AI recognizes your brand, not whether it recommends you. New Balance has the highest KG score in the set and a 3.4% recommendation rate.
Track Brand Mentions in ChatGPT, Claude & Perplexity (2026 Tools)
Track your brand in ChatGPT, Claude, Perplexity & Gemini. 45-min manual workflow plus tool comparisons. Step 2 of the 4-step AI visibility audit.
Find Buyer Questions for AI Prompts: 4 Sources (2026)
Mine Reddit, sales calls, support tickets, and Quora for the real questions buyers type into AI. Step 4 of the 4-step AI visibility audit framework.
40 SaaS Brands, Two GPT Models. Only 12 Passed Layer 1.
A 40-brand AI visibility cohort study across gpt-4o and gpt-5.2. Only 30% pass cleanly. The bottleneck isn't the model. It's the Knowledge Graph.
AI Visibility Audit: 15-Prompt LLM Audit Framework (2026)
Audit brand visibility on LLMs in under an hour. 4-step AI visibility audit framework + 15 free LLM prompts for ChatGPT, Claude, Perplexity & Gemini.