Tag
Start here
In our 20 prompt test ChatGPT searched on 8 prompts and answered the other 12 from memory with no sources at all. Here is a six step check that separates a blank result from a real zero, and the agent job that keeps it running.
Tag
We ran 20 prompts across three AI surfaces and tracked where the sources came from. Five prompts found 31 percent of the source universe, ten found 54, and the twentieth still added 14 new domains. There was no plateau.
Same 20 prompts, three AI surfaces, one day. Gemini cited a median of 47 sources per answer, Perplexity 20, and ChatGPT cited nothing on 12 of 20 prompts while still naming vendors. Here are the two signals a visibility checker has to keep apart.
We sent one prompt to the same model six times with web search on. Across the six answers, 18 distinct URLs were cited and not one of them appeared in all six runs. Here is what an AI visibility tracker has to record if it is going to mean anything.
PageSpeed Insights now has an Agentic Browsing category sitting next to Performance and SEO. This walkthrough shows how to run it, read the fractional score correctly, and fix the six checks it runs.
Meta shipped Muse, a personal AI agent that sends email, books travel, and pays for you. Here is the 20-minute trust review to run before you let it act, and what consumer agents change for AI visibility.
Bing's AI Performance report now shows Citation Share, the slice of AI citations your site actually captures. This walkthrough shows how to read it correctly, segment it with Intents and Topics, and turn the gaps into a weekly optimization loop.
On September 7 Google put interactive charts, simulations, and generated quizzes inside search results. The answer surface is moving from summary to workbench, and that changes what earns a citation. Three shifts, one 30-day response, and the limits worth respecting.
Google now expands AI Overviews into full-page answers that push organic results below the fold. You cannot stop it, but you can decide what stays citable, what you measure, and which pages earn a place inside the answer. A seven-step playbook for holding visibility in the AI-overview era.
Facility managers now ask AI assistants which security firm covers their area. What a local service business needs to be the one named: the trust-record layer, the service-record anatomy and a three-pass plan.
On September 2, 2026, Anthropic open-sourced a shopping agent and a merchant agent on GitHub. Here is what they do, how to run the demos, how to connect them to your own store, and what the release means for AI shopping visibility.
A small business lead travels two halves in 2026: being chosen by an AI answer, then being handled once it arrives. Automation agencies have industrialized the second half; almost nobody is watching the first. Here is the two-half model, the gap it exposes, and a 30-minute way to check where you stand.
A new analysis spanning 75,000 brands quantifies which distribution channels most strongly predict AI search brand visibility. YouTube mentions lead at 0.737 correlation, far outpacing domain rating (0.266) and backlinks (near zero).
