When ChatGPT recommends a running shoe, which of your SKUs shows up?
A brand-level number tells you whether AI names you. Foxish also tracks recommendations per SKU. Import your Shopify catalog, see which of your products AI names on which shopping-intent queries, and audit each PDP for the Product / Offer / Review schema AI actually cites.
Free plan: no credit card. Paid plans: 7-day trial. Or
Curious what a Foxish report looks like first? See live samples for Liquid Death & Magic Spoon →
Is your brand in the AI answer?
Free · 30 secondsWe use one of the actual prompts real shoppers are typing into AI right now — not a marketing question. Live web search enabled. No account required.
3,100+ product categories you can track
See it in 30 seconds
What AI tells your buyers, and what to do about it
Shoppers ask AI what to buy. Foxish shows whether it names you, which of your products it recommends, and what to fix.
Animated walkthrough, 30 seconds, no sound. A shopper asks an AI assistant for the best waterproof winter boots and gets three picks; the store's brand is not among them. Foxish then puts real shopper questions to the main AI assistants and counts how often the brand is named. It scores each product: which one the AI recommends most, which is named without the brand, and which is never recommended. It ends with specific fixes to make and the line: Be the brand AI recommends. All brands and figures shown are illustrative.


Your buyers are already shopping with AI. You can’t see it.
Adobe reported 1,300% year-over-year growth in generative-AI shopping referrals during the 2024 holiday season. ChatGPT has 1B+ weekly active users. When buyers ask AI to recommend products in your category, the answer is already happening — without your brand in the room.
ChatGPT
What’s the best everyday Chelsea boot for wide feet under $250?
For wide feet at that price, the most-recommended options are:
- Blundstone Classic #585 — roomy toe box, best all-day comfort
- Blundstone Work Series #491 — wider last, safety-rated
- Thursday Duchess — narrower, but half-sizes help
Your product isn’t in the answer. And there’s no “AI source” row in your Shopify or GA4.
Already happening
Buyers ask ChatGPT, Perplexity, and Gemini for product recommendations every day — long before they ever search Google or land on your PDP.
Invisible to your stack
Shopify, Klaviyo, GA4, and Triple Whale don’t tell you when AI sent the buyer to a competitor. Your “dark funnel” just got bigger.
Compounds every month
AI traffic is growing exponentially. Every month you’re not in the answer is a month your competitors are training the model to recommend them and not you.
See a real Foxish report — before you sign up
We ran sample reports for Liquid Death and Magic Spoon. Same format you’d get for your brand.
Brand-level tells you you're in the conversation. Per-SKU tells you which product is winning.
A brand's mention rate is useful. But buyers don't buy brands, they buy SKUs. When ChatGPT names “Beautycounter Countertime Vitamin C Serum,” the brand and the product both get credit — and if it's not YOUR SKU in that slot, you're invisible even when your brand isn't.
Brand-level mention rate. You're at 19% overall. Now what?
Now you know: SKU-A12 is winning, SKU-B04 needs work, SKU-C22 is invisible. Fix specifically.
Illustrative example. Brand names and figures are not measurements.
PDP-level AEO grades
Every product page graded on Product / Offer / Review / Breadcrumb schema. Missing markup surfaced with copy-paste JSON-LD.
Category Overlap
Which competitors AI mentions alongside each of your collections. Rows = your categories, cols = competitors, cells = mention rate.
AI-referred revenue
Connect Shopify to see AI-referred sessions, orders and revenue by AI source — what shoppers spent, not just whether they clicked.
Sound familiar?
The rules of product discovery have changed. AI is the new shelf space.
what competitors pay
You spend thousands on Google Ads but ChatGPT recommends your competitor for free
reviews, zero AI mentions
Your Amazon reviews are great but AI doesn't know your DTC site exists
on Google, invisible to AI
You rank #1 on Google but AI recommends a different brand entirely
of research starts with AI
Product research is shifting from Google to AI, and your brand isn't showing up
See the impact for yourself
Plug in your numbers. See what AI invisibility is costing you in lost orders per year.
Revenue Impact Calculator
Estimate the revenue at stake when AI doesn't recommend you
Estimated annual revenue at stake
25K
AI searches/mo
3K
missed visits/mo
82
missed sales/mo
Based on 35% of category searches going through AI, with a top-3 recommendation driving ~12% of those as visits. Your actual impact depends on your category and competition.
Free AEO readiness check
See how AI-ready your website is in under 30 seconds
How it works for product brands
Same AI visibility platform, tuned for ecommerce. Product-specific prompts, review site tracking, and content that makes AI recommend your products.

Illustrative examples with fictional brands and data.
Animation, no sound. A store's catalog is imported from Shopify, WooCommerce or a CSV file, and each product starts to receive a score for how often AI assistants recommend it. Figures are illustrative.
From invisible to recommended
Here's what happens when you connect Foxish to your product brand.
Your first scan comes back
Foxish queries ChatGPT, Google AI Overviews, Gemini, Copilot, and Perplexity with hundreds of prompts real buyers use in your product category.
8%
Visibility Rate
#7.4
Avg Position
Neutral
Sentiment
$142K
Revenue at Risk
Nike Pegasus appears in 62% of AI queries. Your brand appears in 8%.
Everything you need to own your AI presence
16 tools across AI monitoring, social listening, analysis, optimization, and reporting.
Monitor
Visibility, ranking, sentiment scores across all AI models
Pick any of 3,100+ product categories and Foxish builds a 400+ prompt library of buyer questions for it
Word-level changes in AI responses between scans
Shopify alone unlocks sessions + orders + revenue by AI source. Pair with GA4/Mixpanel/Plausible for landing-page depth
Reddit, Twitter, and Instagram monitoring with draft responses
Analyze
Head-to-head on every buyer query that matters
Buyer questions where competitors appear and you don't
How AI groups and categorizes your products
Which review sites AI trusts in your category
Optimize
Estimated annual revenue at risk from AI invisibility
6-point audit with copy-paste Product & Review schema
AI-drafted FAQ, comparison, and product page content
Test any buyer query across all AI sources in real time
Report
3 key numbers + 1 action item in your inbox every Monday
Prioritized action plans with completion tracking and score impact
Multi-client, white-label, automated reporting
Built for product brands
Per-SKU AI recommendation rate
A brand-level number tells you whether AI names you. Foxish also tracks per-SKU: which of your products AI recommends, at what rate, on which prompts. Import from Shopify in one click — 5 SKUs on Free, up to 750 on Executive.
PDP AEO audit across your catalog
Foxish samples your product detail pages and grades each for Product schema, Offer, Review, and Breadcrumb markup — the fields ChatGPT and Perplexity cite when they name your SKU. Missing-schema PDPs are surfaced with a copy-paste JSON-LD generator.
Category Overlap Matrix
For each of your product lines, see which competitors AI mentions alongside you. Rows are your categories, columns are competitors, cells are mention rate — the ecom equivalent of a market-share map, but for AI shelf space.
AI-referred revenue via Shopify
Connect Shopify and see AI-referred sessions, orders, revenue, and CVR — no GA4 install required. Close the loop from a ChatGPT recommendation to your bank account.
Get AI to recommend your SKUs
Foxish tells you exactly why competitors get recommended: their Product schema, Reddit threads, YouTube reviews, marketplace listings. Then generates PDP briefs — hero claims, buyer FAQ, structured-data fields — ready to publish.
Monitor social conversations
Track product discussions on Reddit, Twitter, and Instagram — the citation surfaces AI actually pulls from for ecom answers. Find threads where competitor SKUs are discussed but yours aren't, with draft responses ready to go.
Backed by real research
Princeton's GEO study (KDD 2024) tested 9 optimization strategies across 10,000 queries and found that the tactics that work for AI visibility are different from traditional SEO. We incorporate these findings alongside our own data from thousands of scans.
+41%
visibility from expert quotes
+33%
visibility from statistics
-10%
from keyword stuffing
What this looks like in practice
"What are the best running shoes for wide feet under $150?"
Your brand's in the answer — but only one of your three SKUs. Brand-level tools tell you “you're at 33% mention rate here.” Foxish tells you Coastal Runner is winning while Trail Blazer and Studio Flex need Product schema, review-site placements, and PDP copy tuned to “wide feet” language.
Simple, transparent pricing
Start with a 7-day free trial on any paid plan. Cancel anytime.
SKUs tracked is how many products Foxish scores individually for AI recommendation rate. Product categories is how many distinct buyer-intent surfaces we monitor (e.g. “running shoes” + “trail runners”). AI sources tier up with your plan: Free and Starter cover ChatGPT + Google AI Overview, Growth adds Gemini and Copilot (with its shopping carousel), and Pro and above add Perplexity. Growth and above scan daily. All five sources are included from Pro. Claude (Haiku, one web search per prompt) is an optional add-on from Growth up.
Free
Get your baseline — no credit card required
- 5 SKUs tracked
- 1 product category
- 5 search prompts
- Per-SKU AI mention scoring
- Shopify one-click sync
- 2 AI sources (ChatGPT, Google AI Overview)
- Weekly scans
- Weekly digest email
- 1 top action item per week
Starter
One product line, one AI shelf to defend
Everything in Free, plus:
- 15 SKUs tracked
- 1 product category
- 15 search prompts
- Per-SKU AI mention scoring
- Shopify one-click sync
- 2 AI sources (ChatGPT, Google AI Overview)
- Weekly scans
- Weekly digest email
- 1 top action item per week
Growth
For growing catalogs across a few categories
Everything in Starter, plus:
- 40 SKUs tracked
- 5 product categories
- 75 search prompts (split across categories)
- 4 AI sources (adds Gemini and Copilot)
- PDP schema audit + JSON-LD generator
- Daily scans
- AI-referred revenue (Shopify + GA4)
- Claude add-on available (+$62/mo)
Pro
Full-catalog visibility with share-of-shelf mapping
Everything in Growth, plus:
- 100 SKUs tracked
- 6 product categories
- 100 search prompts
- 5 AI sources (adds Perplexity)
- Daily scans
- Per-SKU competitor share-of-shelf matrix
- Community opportunity finder (Reddit, Twitter)
- REST API access
- Shareable reports
- Claude add-on available (+$92/mo)
Business
Daily monitoring for large multi-category catalogs
Everything in Pro, plus:
- 250 SKUs tracked
- 15 product categories
- 250 search prompts
- Daily scans
- AI-generated PDP content briefs
- Claude add-on available (+$219/mo)
Executive
For big-box catalogs spanning the whole store
Everything in Business, plus:
- 750 SKUs tracked
- 25 product categories
- 400 search prompts
- Unlimited team members
- Claude add-on available (+$346/mo)
Start seeing which of your SKUs AI is naming
Import your Shopify catalog, pick your category, get a per-SKU AI-recommendation score by Monday. Free tier tracks 5 SKUs, Executive goes to 750. No credit card for Free.
Free plan: no credit card. Paid plans: 7-day trial. Choose from 3,100+ product categories.

