The Citation Desk.The Citation Desk.Anyone can ask ChatGPT one question. That tells you almost nothing. Real buyers ask across their whole journey — broad questions, feature questions, comparisons, pricing — often across different AI tools, over days. We measure the whole spread. Here's exactly how.
A brand can be the top answer to "best field service software" and completely absent when someone asks "how do I stop losing track of jobs" — and lose that second buyer entirely.
Measuring a single query is measuring one frame of a whole film. So we never score one prompt. We build a structured spread of buyer questions across six intent types, run them across five AI engines, and measure how often a brand appears across the entire spread. That's the only honest measure of AI visibility.
Every panel we build covers these six. A brand can win one and vanish in another, and that gap is where the opportunity lives.
The buyer knows the category and asks broadly. Reveals who owns the default answer.
"best field service software"
The buyer wants a specific capability. Reveals where a brand punches above its overall rank.
"field service software with QuickBooks integration"
The buyer weighs named options. Reveals who gets pulled into head-to-heads and positioned as the alternative.
"ServiceTitan vs Jobber"
The buyer identifies by their trade or vertical. Reveals sub-category leaders and niche openings.
"software for HVAC contractors"
The buyer describes a pain, not a product. Often returns no brand at all — an uncontested lane for whoever shows up.
"how do I stop losing track of jobs"
The buyer filters by cost or company size. Reveals who owns the budget-conscious and small-team buyer.
"best field service software for under 10 users"
ChatGPT, Perplexity, Gemini, Claude and Google's AI Overview. Each answers differently — and that disagreement is one of the most useful findings.
Every prompt runs in a logged-out, unpersonalised session — so we measure the default answer a real buyer gets, not one shaped by history.
We never ask "is Brand X good?" We ask what a buyer asks — and record who the engine volunteers. That's the measurement.
AI answers shift, so every index is a point-in-time snapshot. We report exact numbers — including where engines named no one, which is itself a finding.
Free tools grade whether a page is technically ready to be cited. That's the setup, not the outcome. We measure whether a brand actually gets named when real buyers ask — across the whole journey, across every engine, read and judged one answer at a time — catching the edge cases and spotting the open lanes a scraper never could.
That method is the same whether the category is field service software, language education, or fine jewellery. Only the questions change. The rigor doesn't.
An audit applies the full method to one brand: all six intent types, all five engines, your competitors benchmarked, and a ranked plan to get named more often. Five days, one fixed fee.
Request an audit →