DEV Community: Han

I Built a Free Tool That Scores How Visible Your Store Is to AI Search

Han — Sun, 16 Aug 2026 06:55:45 +0000

I Built a Free Tool That Scores How Visible Your Store Is to AI Search

A few weeks ago I published a checklist for getting found in AI search: structured data, answer-style product copy, FAQs, speed, topical authority. The feedback I kept getting was the same question:

"Okay, but how do I actually check all that without hiring someone?"

So I built a tool that does it for you. Paste any store URL, and it scores the page against the exact signals AI search engines and shopping agents look for. Here's what it checks — and why each item matters.

The scorecard: 100 points, 6 checks

JSON-LD structured data (30 points) — This is the single biggest miss. AI engines parse schema.org markup to understand what your page is: a Product with a price, an Offer with availability, an Organization with a logo. The most common failure I see: e-commerce homepages with zero JSON-LD at all, even when the site's blog has FAQPage markup. Product pages need Product + Offer at minimum. AggregateRating only if you have real reviews — never fabricate.

Title & meta description (15 points) — Your title tag is still the anchor for how both Google and AI engines label your page. It should name the product and the category. "Home" is a wasted title.

robots.txt (10 points) — Shocking how many stores block their own crawlers without knowing it. If robots.txt disallows your product paths, you're invisible to every crawler that respects it — including AI engines.

Canonical tags (10 points) — Duplicate URLs (with-utm, /products vs /products/) split your authority. Canonicals tell engines which version is the real one. Missing canonicals = diluted signals.

Sitemap (20 points) — AI engines and search engines both use your XML sitemap as the entry point to your catalog. If your sitemap is missing, broken, or missing your newest products, discovery slows to a crawl.

SSL (15 points) — The boring one. No HTTPS, no trust, no ranking. Usually fine, but the tool catches it because people forget on staging domains.

What the scores actually look like

I ran real stores through it while building it:

sellerstack.cc — 50/100. Working sitemap, robots, SSL, title, description. Zero JSON-LD on the homepage. It's an e-commerce toolkit site with decent fundamentals, but AI engines can't tell what it is without structured data.
example.com — 25/100. This is the baseline most small stores sit near: nothing structured, no real description, no sitemap worth crawling.

The pattern is consistent: most stores are one structured-data pass away from doubling their AI-search visibility. It's the highest-leverage, most-neglected fix in e-commerce right now.

How to use the results

Don't treat the score as a grade — treat it as a triage list:

0-30: Fix the fundamentals. Sitemap, robots, SSL, title/description. One afternoon of work.
30-60: Add JSON-LD. Product pages first, then homepage (Organization/WebSite schema), then blog posts (Article schema). If you're on Shopify or WooCommerce, the plugin handles it — just verify it's actually rendering, because most of the time it isn't.
60+: You're doing better than 95% of stores. Focus on content: answer-style product descriptions, FAQs, and guides that make AI engines cite you as a source.

Why this matters more every week

AI shopping agents don't browse like humans. They fetch pages, parse them, extract facts, and recommend. A store with clean Product + Offer markup and answer-rich copy gets recommended; a store without it doesn't exist to them. This isn't a ranking tweak — it's whether you're in the game at all.

The checklist I published earlier is the theory. The tool is the practice. Run your store through it, fix what's broken, and you've done more for your AI-search visibility than most competitors will do all year.

If you're a small e-commerce brand trying to stay visible as search changes, this is exactly the kind of stuff I work on. I build tools for sellers navigating these shifts — SellerStack is the e-commerce toolkit I'm building around them.

Your Store Is Invisible to AI Search. Here's How to Fix It

Han — Mon, 10 Aug 2026 06:39:54 +0000

Your Store Is Invisible to AI Search. Here's How to Fix It

AI assistants are becoming the new storefront. When someone asks ChatGPT, Gemini, or Perplexity "best lightweight rain jacket under $100", they get an answer — and the stores that appear in that answer get the traffic. The rest get nothing.

If you run an e-commerce store, this isn't a future problem. It's happening now, and most store owners haven't touched the things that decide whether AI can even understand their product pages.

The good news: you don't need to be a big brand to get found. You need to do the boring technical stuff right. Here's the checklist.

1. Structured data is your entry ticket

LLMs and AI search engines crawl your pages like traditional search engines do. That means schema.org markup matters more than ever.

The most important types for e-commerce:

Product — name, price, currency, availability, brand, image
Offer — price, priceValidUntil, availability (InStock/OutOfStock)
AggregateRating — only if you have real reviews (never fake this)
BreadcrumbList — helps AI understand your site hierarchy

Every product page should have Product + Offer markup. Shopify and WooCommerce have apps/plugins for this, but don't assume it's on — verify with Google's Rich Results Test or Schema.org validator.

2. Write product copy that answers questions

AI systems extract answers from your content. If your product page only says "Premium quality, best price!", there's nothing to extract.

Structure your product descriptions like you're answering a customer's actual questions:

What problem does this solve?
What are the specs (weight, size, materials)?
What's it compatible with?
How does it compare to alternatives?

One product page that clearly answers 10 real questions beats 50 pages of marketing fluff. AI engines favor pages with comprehensive, factual, well-structured content.

3. Your FAQ pages are suddenly your best SEO asset

FAQ content is gold for AI search. LLMs love Q&A format because it maps directly to how people ask questions.

Turn your support emails, chat logs, and returns data into FAQ sections. Answer the actual questions people ask — not the ones you wish they'd ask.

4. Speed and mobile aren't optional

AI search engines inherit ranking signals from traditional SEO. A slow, mobile-unfriendly site is a liability in every channel. Compress images, use a CDN, keep your Core Web Vitals green. This is table stakes.

5. Build topical authority with content

AI engines trust sources that consistently cover a topic. A store that publishes genuinely useful guides about outdoor gear becomes the source AI pulls from for outdoor gear questions.

This is a compounding play: every article, guide, and FAQ you publish increases the chance AI cites you — which brings traffic, which builds authority, which makes AI trust you more.

What to do this week

Run your homepage through Google's Rich Results Test — is Product markup present?
Add/verify Product + Offer schema on your top 10 products
Rewrite one product description to answer 5 real customer questions
Create one FAQ page from your actual support conversations

None of this requires a big budget. It requires treating your product pages as answers, not ads.

AI search isn't coming for your traffic — it's already deciding who gets traffic. Make sure it can find you.

If you're a small e-commerce brand trying to stay visible as search changes, this is exactly the kind of stuff I write about. Follow along or check out SellerStack — an e-commerce toolkit for sellers navigating these shifts.

AI Shopping Agents Are Already Buying From Your Store. Here's How to Get Found

Han — Sun, 02 Aug 2026 14:13:43 +0000

AI Shopping Agents Are Already Buying From Your Store. Here's How to Get Found

You open your analytics dashboard and traffic from Google is down 18% this week. Your ads are running. Your SEO hasn't changed. So where did the visitors go?

They asked an AI agent to shop for them instead.

This isn't a prediction anymore. As of mid-2026, AI shopping agents are becoming the default way a growing slice of consumers buy online. PayPal and Commerce research found 64% of UK consumers already want to use agentic AI for shopping. Samsung partnered with Glance to bring agentic commerce to smart TVs. During this year's Prime Day, major retailers raced to make their product data machine-readable so AI agents would recommend them over competitors.

If you still treat AI as just a tool for writing product descriptions or generating ad creatives, you're about to miss the biggest shift in online shopping since mobile.

What Agentic Commerce Actually Means

Agentic commerce is when an autonomous AI agent shops on behalf of a person. Instead of a customer typing a query into Google, clicking five results, comparing specs, reading reviews, and checking out, they tell their agent: "find me the best wireless noise-cancelling headphones under $200 and order them with my saved card."

The agent searches multiple stores, compares products against the buyer's stated preferences, reads reviews, checks return policies, and completes the purchase. The customer never visits your website.

Google's Shopping AI, Amazon's Rufus, and ChatGPT's shopping capabilities already do versions of this today. What changed in 2026 is trust: consumers are letting agents complete real transactions, not just research.

Here's the critical difference. Agents don't browse like humans. They don't scroll your homepage, absorb your brand story, or wander through your category pages. They consume structured data: product feeds, pricing, specifications, reviews, availability, shipping, and return policies. If your product data isn't clean and machine-readable, you are invisible to agents. Full stop.

Why This Is Urgent for Sellers Right Now

The market signals are hard to ignore:

AI search traffic is growing while organic stagnates. Adobe's research from April 2026 shows AI-driven search traffic to retail sites is growing, but most retail sites lag badly in AI search visibility. Consumers are using AI search; stores haven't optimized for it.

Retailers are actively courting bots. An Observer report from June 2026 documented how retailers optimized for AI agents during Prime Day, making product data machine-readable and pushing to get recommended by AI shopping assistants.

The big platforms are all-in. Amazon began selling its AI shopping technology to other retailers in May 2026, signing partners like Kate Spade. Samsung's TV partnership brings agentic commerce to the living room. PayPal is investing in agentic checkout infrastructure. When Amazon, Samsung, and PayPal pivot at the same time, that's a structural shift, not a fad.

The old playbook of optimizing keywords, running ads, and building backlinks still matters. It's just no longer sufficient. You now also need to be discoverable and recommendable by AI agents.

How Agents Evaluate Your Products

Agents don't care about beautiful photography or emotional brand storytelling. They evaluate structured signals:

1. Data Completeness and Accuracy

Your product feed must include detailed specifications (dimensions, materials, compatibility), accurate pricing including promotions, real-time inventory, shipping costs and timeframes, return policy, and warranty info.

Missing data doesn't just hurt your ranking. Agents filter out products with incomplete information entirely. One missing field can make you invisible.

2. Review Quality and Sentiment

Agents scan reviews at scale for sentiment patterns, praise points, and recurring complaints. Products with more reviews, higher average ratings, and detailed positive feedback get recommended more often. Your review generation strategy is now part of your agentic visibility strategy.

3. Total Price Competitiveness

Agents compare total cost: product price plus shipping. A competitive base price with heavy shipping can get you deprioritized. Transparent, competitive total pricing wins.

4. Trust Signals

Generous return policies, clear warranty info, verified seller badges, and responsive customer service all influence whether an agent recommends you. Agents are optimizing for customer satisfaction, and trust signals predict it.

An Actionable Checklist

Here's what to do this month, in priority order:

Audit your product feed. Export it and check for missing specs, stale pricing, or wrong inventory flags. Fix the data before touching anything else.
Add structured data to your product pages. Schema markup (Product, Offer, Review, AggregateRating) tells both search engines and agents exactly what each product is.
Standardize pricing and shipping. Make sure every platform you sell on shows consistent, accurate totals.
Build a review engine. Post-purchase email sequences that convert buyers into reviewers are cheap and compound directly into agent recommendations.
Publish your return policy clearly. Both on the product page and in your feed.

The Verdict

Agentic commerce is not a threat to good products. It's a filter that rewards sellers with clean data, honest pricing, and real social proof, and it punishes sellers who rely on brand gloss alone.

The sellers who treat their product data as a first-class asset will capture the traffic everyone else loses to agents. Start with the feed audit. It costs nothing but an afternoon, and it's the single highest-leverage move you can make right now.

First published as a test article. Data referenced: PayPal/Commerce agentic shopping research, Adobe AI search visibility report (April 2026), Observer Prime Day coverage (June 2026).