AI Recommendation Audits

Your score is what AI can read.
This is what AI recommends.

The free scan grades the signals on your site. Our audits run the real category queries your shoppers ask ChatGPT, Claude, Perplexity, and Gemini, and show whether they name you, rank you, or recommend a competitor instead.

See the auditsOne-time. No subscription. Report in about 24 hours.
01 Read vs recommend

You fixed what AI can read. Now see what it recommends.

The free score

On-site signals you control.

  • Schema, crawlability, content depth, trust signals
  • Graded from your live store in about a minute
  • Tells you what an assistant can read about you
The audits

Off-site authority you earn.

  • Who the web says you are, who the models cite and name
  • Measured by running the queries shoppers actually ask
  • Tells you what the assistants recommend
02 Why off-site is the half that matters

AI doesn't invent recommendations. It repeats them.

When an assistant names a brand, it's echoing what its sources already say. That's why a flawless store can still go unnamed, and why this is the half worth measuring.

01

Your website alone is not enough anymore

AI assistants do not just read your store. They pull from the sources they trust most: big reference sites, public data, and the places people actually talk, like Reddit and YouTube. So you can rank on page one of Google and still never show up in an AI answer. If the wider web does not back up your brand, the assistant points shoppers to a competitor it knows better.

02

Getting mentioned is what moves you

Researchers at Princeton and Georgia Tech tested this directly. Pages backed by real citations, stats, and quotes showed up 30 to 40 percent more often in AI answers. Being talked about, with receipts, is what gets you surfaced.

03

Recommendations follow demand

What AI names tracks how often people search for you and where you show up across the web. It has little to do with how clean your theme is. The audit shows where you stand on the signals that actually decide it.

03 What the audit measures

Four signals behind every recommendation.

01

Recommendation rate

How often each assistant names you across the real category queries shoppers run, and where you rank when it does.

02

Knowledge-graph presence

Whether Wikipedia, Wikidata, and the knowledge panels the models lean on actually know your brand exists.

03

Citation & source authority

Which sources the assistants cite for your category, and whether your brand shows up in any of them.

04

Sentiment & framing

When you are mentioned, how you get described, and which competitors you get measured against.

04 What you get back

Every query, every assistant, in one grid.

A brand-by-query-by-assistant matrix: where you are named, where you rank, and where a competitor gets recommended instead.

Category queryChatGPTClaudePerplexityGemini
best clean beauty brand#6
affordable vitamin C serum#3#4#5
cruelty-free skincare#8
natural anti-aging products#2
7 of 20 prompts named
4.6 average rank
0 first-place results
+16 more queries in the full report, each with the competitor named instead.
05 How it works

Three steps. No call, no install.

01

Submit your store

One URL. No tag to install, no access to grant.

02

We run the real queries

Real category prompts across ChatGPT, Claude, Perplexity, and Gemini, repeated to catch the variance.

03

You get the report

Your recommendation rate, the misses, who got named instead, and a prioritized plan.

06 Choose your audit

Start anywhere. Start now.

One-time. One store. No subscription. Report straight to your inbox in about 24 hours.

AI Recommendation Check
$19.99

A fast, focused answer.

Your top category query asked across ChatGPT, Claude, Perplexity, and Gemini. Do you make the list, and who does? The fastest way to see where you stand.

  • 1 category query, all 4 assistants
  • Recommendation result: named or not
  • Who gets named instead
Get the Check
AI Recommendation Blueprint
$500Per store

For brands serious about owning their category.

Everything in the Audit, plus deep competitor teardowns, a 60-day re-check to prove the lift, and a working session to map the strategy.

  • Everything in the Audit
  • Deep competitor teardowns
  • 60-day re-measurement
  • Strategy working session
Get the Blueprint

Built for brands already serious about growth. Most stores start with the Audit.

Secure checkout. Full report in about 24 hours, straight to your inbox. No sales call.

07 Questions

Before you start.

How is this different from the free score?

The free score grades on-site signals: schema, crawlability, content, trust. It measures what an assistant can read about you. These audits measure the opposite end: whether those assistants actually recommend you when a shopper asks, which is driven by off-site authority you do not control from your theme.

How do you query the assistants?

We run real category prompts across all four assistants, several times each, and record whether your brand is named, where it ranks, and who gets recommended instead. Repetition smooths out the run-to-run randomness so the numbers mean something.

How long does it take?

About 24 hours. The queries run on our side, then we compile the report and the plan and email it to you. No call, no scheduling.

What does a fix look like?

Concrete and sequenced: the knowledge-graph entries to claim, the citation sources worth earning, and the framing gaps to close, ordered by how much each is likely to move your recommendation rate.

Do I need to run the free scan first?

No. You can go straight to an audit. The free scan tells you what AI can read about your store; the audits tell you whether it recommends you. They are complementary, but neither requires the other.