Blog · Topic
Analytics
Turning collected answers into numbers that hold up: mention rate, citation share, sample size, volatility and the reports built on top of them.
Articles
ChatGPT, Gemini and AI Mode name different leaders in half our tracked categories
We asked three AI engines 182 buyer-intent prompts across 10 software categories. In five categories the engines name a different leader; the October 2026 data.
How much AI answers change between runs: 30 questions, 10 runs each
We asked ChatGPT and Google AI Mode the same 30 buying questions ten times each. Two thirds of the brands ChatGPT named showed up only some of the time. How many runs you need for a number you can report.
AI answer volatility: why one check is an anecdote
The same prompt can name you today and a competitor tomorrow. How to measure answer variance with rates, standard errors and sample sizes.
AI monitoring cost planning: estimate the bill up front
Size prompts, markets, engines, cadence and add-ons into a monthly dollar figure, with worked arithmetic before your first run.
An AI visibility report clients and executives will read
A monthly AI visibility report template: three metrics, what changed, an action list, and how to compute each number from stored API responses.
GEO metrics vs SEO metrics: what maps and what doesn't
Position, impressions and CTR don't survive inside AI answers. The metric translation, formulas for each GEO metric, and where each engine returns it.
Sampling vs. census in AI-answer monitoring
How many runs per prompt are enough: intervals, sample sizes for detecting change, and when to collect everything instead of sampling AI answers.
AI grounding by engine: which index each assistant uses
What each AI engine documents about its search index and citations, what it doesn't, and a method to measure how its citations overlap with Google.
AI rank tracking: what rank means inside an AI answer
Six definitions of rank inside AI answers, how to aggregate when you are absent, and the statistics and sample sizes behind a trustworthy number.
Best AEO tools for 2027, organized by the job you need done
AEO tools grouped by job: tracking answers, deciding what to change, AI crawler access and first-party data, with verified prices and selection criteria.
Best AI visibility tools for 2027: GEO platforms compared
Profound, Peec AI, Otterly.ai, Scrunch, AthenaHQ, Semrush and Ahrefs Brand Radar compared on engines, prompts and entry price, plus the raw-data API route.
Best competitive intelligence tools for 2027, by the job
Crayon, Klue, Kompyte, Owler, Similarweb, Semrush, Ahrefs and SpyFu by job, with prices checked on vendor pages, and when to build on raw SERP and AI data.
Competitive analysis template with a source for every row
A competitive analysis template covering positioning, pricing, search, AI answers, news and ads, with a source per row, scoring and refresh cadence.
18 competitive intelligence examples from public data
Concrete competitor questions, each with the endpoint, an example query or prompt, the exact response field to read, and what the data cannot tell you.
SERP and AI answer data in BigQuery: schema and queries
Load SERP and AI answer results into BigQuery: a partitioned schema with nested results and JSON, free batch loads, streaming costs and example SQL.
How to measure the traffic AI engines send you
AI referral traffic is partial and noisy. How to identify it in analytics, join landing pages to cited URLs, and test what it can't show.