Under the hood

Inside the engine

CitoRank doesn't guess what AI thinks of your store — it asks. Here's exactly what happens between "Install" and your first plan for winning the AI recommendation — not just a report that a rival won it.

Step 01

Deep store analysis

We connect securely to your Shopify catalogue via the official Shopify API. We pull every product, description, price and variant, then scan the rendered JSON-LD on each product page — so we see exactly what an AI model "sees" when it reads your store, not just what your admin says is there.

Shopify Admin API Product catalogue JSON-LD
Step 02

Reverse-engineer buyer intent

Our prompt engine auto-generates hundreds of variations of the real questions shoppers ask before they buy — no CSV imports, no manual keyword research. We simulate every angle of the funnel:

  • Discovery: "best camping fridge under £300"
  • Use-case: "best camping fridge for an off-grid campervan"
  • Specifics: "is the CFX3 25 compatible with a 100Ah battery?"
Prompt engine Per-model planner Intent analysis
Step 03

Live LLM interrogation

We don't simulate — we ask. CitoRank fires every prompt at ChatGPT, Claude and Gemini through their official APIs and captures the exact citations each model returns. On our Balanced and Comprehensive tiers, ChatGPT and Claude use live web search to ground their answers in Trustpilot, Reddit, news and brand sites; Gemini answers from its model knowledge — so what you get back is evidence, not hallucination.

OpenAI Anthropic Google Gemini web_search
Step 04

The authority gap

If you aren't ranking, who is? CitoRank identifies the exact sources the AI models cite when they recommend your competitors — the Trustpilot pages, the Reddit threads, the listicles, the news mentions. We measure your share of voice against theirs across the wider web — the off-site authority no catalogue feed can syndicate for you — sentiment included, then prioritise the gaps by impact.

High Example recommendation: pitch your fridge to outdoorsmagic.com's "best camping fridges" round-up — that one source backs 3 of the 5 answers we tracked for this keyword, and two competitors are on it while you aren't.
Competitor analysis Source ranking Citation tracking
Step 05

Schema & content scorecard

Finally, we audit your technical SEO from an AI's perspective. If ChatGPT can't parse your price, availability or product identifiers, it won't recommend you with confidence. We score the live JSON-LD on every product page against Google's structured-data guidelines and flag the exact fields you're missing or breaking — so your developer (or theme) knows precisely what to fix.

High Example recommendation: add AggregateRating to your Product JSON-LD — your reviews are invisible to the models, so ChatGPT quotes your competitor's Trustpilot score instead of yours.
JSON-LD scoring Schema.org Microdata

See whether AI recommends you — or a rival

Install in one click. Your first audit runs the same day.

Install on Shopify See pricing