Surfer is the AI’s top pick in 1 of 16 measured B2B categories and invisible in 14. Even where it leads, the engines do not agree: at best 1 of 8 name it first.fragile
Where Surfer is a top pick, on average only 13% of the engines that answered agree on it. This report grades that as fragile. A single-engine tracker would show you one of those views and call it your score. In this dataset, across all 16 categories, no tool is the unanimous pick of all 8 engines. The disagreement is not noise to average away; it is the real state of your AI visibility, and it is what every single-number tracker quietly hides.
Each row is a category where at least one engine ranks Surfer first. The dissent column is the honest part: the engines that recommend a competitor instead, on the same date.
| Category | Engines that agree | Engines that pick something else |
|---|---|---|
| GEO / AI visibility | 1 of 8 | Llama 3.3 70B via Groq picks Ahrefs Cohere Command-A picks Ahrefs Gemini 2.5 Flash-Lite picks Ahrefs Perplexity picks Siftly ChatGPT picks Profound Llama 4 Scout picks Ahrefs Qwen 3.6 picks None |
These are categories Surfer genuinely competes in, where the engines list it but put someone else first. This is the gap a buyer asking an AI sees before they ever reach your site.
| Category | Your rank | Ranked ahead of you |
|---|---|---|
| GEO / AI visibility | #7 | Ahrefs, BrightEdge, SEMrush |
| SEO tools | #3 | Ahrefs, SEMrush |
In these 14 categories, no measured engine surfaced Surfer at all. Some may be outside your market; the ones inside it are pure lost recommendation share.
Across all 16 categories, no single tool is the #1 pick of every engine. The "AI visibility" number a single-engine tracker shows Surfer is one engine’s view on one date; this report shows all 8.
This report is one dated capture, given to you at no cost. AI recommendations move every run and every week. The AI Visibility Tracker keeps score for you:
Each of 8 AI engines (Llama 3.3 70B via Groq, Cohere Command-A, Gemini 2.5 Flash-Lite, Perplexity, ChatGPT, Llama 4 Scout, GPT-OSS 120B, Qwen 3.6) answered the same buyer questions across 16 B2B and go-to-market software categories. Every recommendation is a recorded model output, canonicalized and scored by position-weighted frequency. Capture window 2026-06-19 to 2026-07-08.
The full dataset is published open under CC-BY-4.0 with DOI 10.5281/zenodo.20767878. This report states what the engines returned on the capture date. It does not judge product quality, and no vendor can buy a better number; the rankings are unpurchasable by design.