AI visibility for Shopify

ChatGPT is recommending your competitors. You're invisible.

Shopify auto-enrolled your store into ChatGPT Shopping and Google AI Mode. Enrollment isn't visibility. If your catalog can't tell an agent what your products are, it recommends someone else — and you never see it happen, because there are no impressions to report.

30 seconds. No signup.

LIVE RETRIEVAL TEST
CHATGPTPERPLEXITYGOOGLE AI MODECOPILOTGEMINI

SHOPPER ASKS

> best magnesium supplement for sleep

ASSISTANT RECOMMENDS

1Competitor Aglycinate, third-party testedCITED
2Competitor B400mg, no fillersCITED
3Competitor Csubscription, sleep-specific blendCITED
—Your storenot retrievable — category missingABSENT

WHY YOU WERE SKIPPED

PRODUCT TAXONOMYmissing
METAFIELDS5 of 9 empty
GTIN / MPNnot set
PRODUCT JSON-LDpartial
Weighted retrieval readiness34
●best magnesium supplement for sleep●organic cotton crib sheets under $80●which running shoe for flat feet●gift for someone who loves coffee●fragrance-free moisturiser for eczema●durable dog harness for pulling●best standing desk for small apartments●non-toxic cookware that lasts
The recommendation gap / 2026

The channel is already moving.

8x

AI-driven traffic growth to Shopify stores, Q1 2026 YoY

13x

increase in orders from AI-powered search

2x

rate at which new buyers convert through AI channels

Adobe found AI-referred shoppers bounce 33% less and convert 31% more. 94% of marketing leaders are increasing AI-search budgets this year. The merchants winning these recommendations right now are the ones who fixed their catalog first — and catalog fixes compound. Every week a competitor is indexed correctly and you aren't is a week of recommendations you don't get back.

The blind spot

You can't see the traffic you're not getting.

There's no impression to report, no rank to track, no analytics event to investigate. The first signal is flat revenue with no obvious cause — which is why most merchants find out six months late.

SHOPIFY ANALYTICSLAST 90 DAYS

AI CHANNEL IMPRESSIONS

—

A flat analytics line with no reported AI impressionsno datano data

Nothing in Shopify admin reports an AI recommendation you didn't win. The channel is invisible by construction.

CATALOG SCANRUNNING
SKUCategoryFieldsStatus
SLP-001missing5 emptyHIGH
CLM-114generic3 emptyMED
MAG-402set1 emptyLOW
SLP-002missing6 emptyHIGH
RST-088generic4 emptyMED

Two ways in.

I run a Shopify store

Find out whether AI assistants can see your products, which competitors are taking your recommendations, and exactly what in your catalog is causing it.

Run a free scan

I run an agency

Add AI visibility to your service offering without building the tooling. White-labeled reports under your brand, unlimited scans across your client book, priority turnaround.

See white-label plans
What gets audited

Three layers. Forty-plus checks.

Most stores fail at the first layer and never find out, because nothing in Shopify admin tells them. The audit starts where AI systems do: with the data they can retrieve. Then it follows each missing signal through your site and your channels, until the exact fix is clear.

CATALOG LAYER22 CHECKS
  • Missing or generic Shopify taxonomy categories — the single most common reason a store gets skipped entirely
  • Empty metafields on decision attributes: material, dimensions, use case, compatibility, ingredients
  • Missing product identifiers — GTIN, MPN, Google product category
  • Catalog Mapping gaps: data trapped in metaobjects, tags, or title delimiters where Catalog can't source it
  • Naming inconsistency across variants (“100% cotton” vs “Cotton Blend”) — the top trigger for AI hallucinating your specs
  • Variant completeness and per-variant media coverage
  • B2B-only or gated products leaking into agentic channels
AGENT-READABLE LAYER12 CHECKS
  • /llms.txt, /llms-full.txt, /agents.md — what they actually expose
  • Product JSON-LD validity and completeness
  • Knowledge Base coverage: returns, shipping, sizing, FAQ, brand voice
  • Crawler access — whether AI agents are being blocked at the edge
CHANNEL CONFIG8 CHECKS
  • Shopify Catalog eligibility status
  • Google & YouTube channel connection (required for Google AI Mode and Gemini surfaces)
  • Per-channel configuration and market coverage
The method

Most 'AI audits' are a screenshot.

AI answers vary between askings and between phrasings. Ask the same question twice and you can get two different brand lists. A single screenshot of one ChatGPT reply proves nothing — which is why most of what's sold as an AI audit is a screenshot and a sales call.

SAME QUESTION, ASKED 5×5 DIFFERENT ANSWERS
RUN 01
Comp AComp CComp F
RUN 02
Comp BComp A
RUN 03
Comp CYOUComp B
RUN 04
Comp AComp DComp B
RUN 05
Comp BComp C
One screenshot would have caught exactly one of these.
  1. 01

    PROMPT BASKET

    25–40 real buying questions in your category, drawn from how customers actually ask, not keyword tools.

  2. 02

    SAMPLED BATTERY

    Every question asked repeatedly across phrasings, so results carry confidence ranges instead of anecdote.

  3. 03

    FIVE ENGINES

    ChatGPT, Perplexity, Google AI Mode, Copilot, Gemini.

  4. 04

    CITATION SHARE

    How often you're named versus each competitor, measured not guessed.

  5. 05

    COMPETITOR GAP MAP

    Which three brands own your queries, and precisely what's in their catalog that isn't in yours.

  6. 06

    90-DAY RE-MEASURE

    The same battery re-run after fixes, so the improvement is proven rather than claimed.

Sample data

Here's what you actually get.

An audit isn't a scorecard. It's a working map from what an AI can't retrieve to the exact data change that makes it legible.

OUTGROWW / VISIBILITY REPORTSAMPLE DATA

Visibility score

WEIGHTED RETRIEVAL READINESS

Sample visibility score: 34 out of 10034SAMPLE SCORE / 100
01

VISIBILITY SCORE

A single number out of 100, weighted by how much each gap actually affects retrieval. 34 means agents can parse roughly a third of what they need to recommend you.

02

CITATION SHARE

Across 40 questions and 5 engines. The gap between 4% and 31% is not a branding problem — it's a data problem.

03

PRODUCT GAP TABLE

Every product, ranked by how much its gaps cost you. Usually 20% of the catalog causes 80% of the invisibility.

04

THE QUESTIONS

You see the actual questions, the actual answers, and the actual brands named ahead of you. Nothing summarized away.

05

FIX LIST

Ordered by impact per hour of work. You can hand this straight to a developer, or I implement it.

06

90-DAY RE-MEASURE

The same battery, re-run. This is the number that tells you whether any of it worked.

Compare

A screenshot is not a measurement.

 OUTGROWWSCREENSHOT AUDITSEO AGENCYDIY
Visibility score you can re-measureYESNONONO
Full prompt battery, sampled repeatedlyYESNOsometimesNO
Five engines, not oneYESNO1–2NO
Confidence ranges instead of one answerYESNONONO
Product-level gap table, every SKU rankedYESNONONO
Shopify Catalog & metafield specificsYESNOvariesmanual
Implementation done for youYESNOYESNO
90-day re-measure to prove the resultYESNONONO
Glossary

The vocabulary, in plain English.

This space invented a lot of acronyms very quickly. Here's what they actually mean.

Outgroww AI visibility glossary
AGENTIC COMMERCEShopping where an AI assistant does the searching, comparing, and recommending on the shopper's behalf, instead of the shopper browsing a store directly.
AEO (ANSWER ENGINE OPTIMIZATION)Optimizing to be the answer an AI gives, rather than a link on a results page.
GEO (GENERATIVE ENGINE OPTIMIZATION)Broadly interchangeable with AEO. Optimizing for inclusion in AI-generated responses.
SHOPIFY CATALOGThe Shopify system that syndicates your product data to AI surfaces like ChatGPT Shopping and Google AI Mode. Most stores are enrolled automatically and assume that means they're visible.
PRODUCT TAXONOMYShopify's standardized category tree. A missing or generic category is the most common reason an agent skips a product.
METAFIELDA custom data field on a Shopify product. Where attributes like material, dimensions, and use case live. Usually empty.
CATALOG MAPPINGThe configuration that tells Catalog where to find your attribute data. Data in the wrong place is invisible data.
GTIN / MPNGlobal Trade Item Number and Manufacturer Part Number. Identifiers that let an agent match your product to a known item.
JSON-LDStructured data embedded in your page that describes your products in a machine-readable format.
/LLMS.TXTA root file describing your site for large language models, analogous to robots.txt. Shopify serves one automatically.
CITATION SHAREThe percentage of relevant AI answers in which your brand is named. The core metric of AI visibility.
PROMPT BASKETThe defined set of buying questions an audit tests against. The quality of the basket determines the quality of the audit.
HALLUCINATION RISKThe likelihood an AI states something incorrect about your product. Rises sharply with inconsistent catalog data.
GROUNDINGWhen an AI's answer is anchored to retrieved source data rather than generated from memory. You want to be the grounding.

Pricing.

FREE SCAN

$0

instant
  • Automated catalog check across your full product set
  • Visibility score out of 100
  • Your top 3 gaps named specifically
  • No signup, no card
MOST POPULAR

VISIBILITY REPORT

from $299

3 days
  • Everything in Free Scan
  • Full prompt battery, 25–40 questions across 5 engines
  • Competitor citation share with confidence ranges
  • Product-level gap report, every SKU ranked
  • Prioritized fix list with effort estimates
  • 30 minutes of walkthrough over call or Loom

AUDIT + IMPLEMENTATION

from $1,200

7 days
  • Everything in Visibility Report
  • I implement the catalog, metafield, and schema fixes directly
  • Catalog Mapping configured properly
  • llms.txt and JSON-LD corrected
  • 90-day re-measure included, so the result is proven

MONITORING

$199/mo

monthly
  • Monthly re-scan of the full battery
  • Drift alerts when your score drops
  • New-competitor alerts when a brand starts taking your queries
  • Quarterly review call

Tell me your store URL. I'll tell you what I find.

If your score is fine, I'll tell you that and we're done. I'd rather lose the sale than sell you an audit you don't need.

A few practical questions.

What does Outgroww audit?

Outgroww reviews three layers of Shopify AI visibility: catalog data, agent-readable site data, and channel configuration. The audit checks product taxonomy, metafields, identifiers, catalog mapping, structured data, crawler access, and relevant channel eligibility.

What is an AI visibility score?

The score is a number out of 100 weighted by how much each catalog, machine-readable data, and channel gap affects an AI assistant's ability to retrieve and recommend a product.

What is included in the free scan?

The free scan is an automated check across the full product set. It returns a visibility score out of 100 and identifies the store's three most important catalog gaps. It requires no signup or card.

How does a full visibility report measure AI visibility?

A full report uses a basket of real buying questions, sampled repeatedly across phrasings and across ChatGPT, Perplexity, Google AI Mode, Copilot, and Gemini. It measures citation share against competitors and reports confidence ranges rather than relying on a single answer screenshot.

Can Outgroww implement the recommended fixes?

The Audit + Implementation service includes direct implementation of catalog, metafield, schema, catalog mapping, and llms.txt fixes, followed by a 90-day re-measure.

Does Outgroww offer plans for agencies?

Yes. Agency options include white-label reports, unlimited scans across a client book, priority turnaround, and a per-client option for agencies testing the service.

Find out what an agent sees. In 30 seconds.

No signup, no card. If your catalog is already in good shape, the scan will tell you that.

Run a free scanSee pricing