Case Study: One Product Title Rewrite, 40-Point AI Readiness Score Jump

By Steve Merrill | June 3, 2026

During a client audit last week, I ran a single product through our AI readiness scoring system. It scored 57, below the average we see across Shopify stores. Not the worst score I've seen. But firmly in the range where AI shopping assistants are going to pass it over in favor of better-documented products.

I rewrote one field. Just the title. No new description, no new images, no review campaigns. Ran the score again: 97.

That's 40 points from one change. And it's not a one-off, I've now seen this pattern across enough audits that it's become the first place I look on any new store.

What Was the Original Title?

I'll use a close analog to preserve client confidentiality. The original title was something like: "Classic Chino Pants, Navy, Size 32"

To a human scanning a product grid, that's fine. Descriptive enough. The color and size are right there. But to an AI shopping assistant trying to answer a question like "What are the best slim-fit work pants for a tech office dress code?", that title is nearly useless. No brand. No use case. No audience context. Just a category label with some attributes.

What Was the Rewritten Title?

The rewrite followed a specific formula we've tested across hundreds of products. It became: "[Brand] Men's Slim Chino for Business Casual, Navy Cotton Stretch, Sizes 28-38"

Same product. Different information architecture. The brand name is now present. The fit type is explicit. The use case ("business casual") is spelled out in the title itself. The size range replaces the single-size attribute, which serves catalog-browsing customers better anyway.

That one structural shift moved the score from 57 to 97.

Why Does Title Structure Affect AI Score This Much?

Product titles are weighted heavily in AI readiness scoring because they're the primary field AI shopping systems use when evaluating product relevance to a query. The description matters too, but AI systems frequently parse the title first to determine whether to examine the rest of the product record.

A title without a brand name means the AI can't confidently associate the product with a brand entity. That breaks brand-query recommendations: when someone asks "what are [Brand]'s best work pants?" the product might not surface even if everything else is perfect.

A title without a use case means the AI can't match the product to conversational queries. Search Engine Journal's research on AI Overviews product visibility confirms that products with use-case context in their titles appear significantly more often in AI-generated shopping recommendations than spec-only titles.

A title with only a single size variant limits product reach. AI systems prefer products with clear availability breadth, it signals that the product is actively stocked and broadly applicable.

The Formula Behind the Rewrite

Here's the exact four-part structure we apply. Every element earns points in the AI scoring system:

Part 1: Brand Name. First word in the title. Non-negotiable. The brand entity connection is foundational, every AI system I've tested uses it to connect products to brand queries, brand reputation signals, and brand-level review data.

Part 2: Product Name with Specificity. Not just "chino pants." "Men's slim chino" or "women's cropped cargo", the variant type matters. Specificity allows the AI to match the product to a narrower, higher-intent query.

Part 3: Use Case or Audience Anchor. A short phrase that places the product in a real context. "for business casual," "for outdoor workwear," "for summer events," "for marathon training." This is the phrase that connects your product to the actual questions buyers are asking AI shopping assistants.

As Google's product structured data documentation notes, products that clearly describe their application and audience are better positioned for both Shopping and organic recommendations.

Part 4: Key Specs. Material, size range, primary color or colorway, quantity if relevant. These serve Google Shopping matching and filter-based browsing. Don't drop them, just put them at the end.

How Far Do These Gains Compound?

The 40-point jump from one title is striking. But the compounding story is what really matters.

When a store rewrites its top 50 products using this formula, the aggregate AI readiness score typically shifts by 8-12 points. That moves a lot of stores from the 45-65 average range into the 70+ tier where AI recommendation frequency starts to climb meaningfully. Schema.org's product specification provides the technical framework, but the title structure is the human-readable layer that AI systems actually read when forming product recommendations.

The clients I've seen do this consistently, building a library of strong titles across their catalog, compound away from competitors who are still running keyword-only title structures. The gap isn't visible week to week. But at six months, it's clear.

What Should You Do First?

Export your product catalog. Sort by traffic or revenue. Take your top 20 products and audit each title against the four-part formula. How many are missing the brand name? How many have no use case? How many are just a category name and a color?

Rewrite those 20. Run them through an AI readiness check. The score movement will tell you exactly how much headroom you have, and whether a full catalog rewrite is worth the effort.

In most stores I audit, it is. By a significant margin.

Frequently Asked Questions

What makes a product title score high on AI readiness?

AI readiness scores reward product titles that contain a brand name, a natural-language use case, a qualifying audience or context, and key product specifications. Titles that are only spec-strings or category labels score poorly because AI systems can't extract enough meaning to confidently recommend the product.

How much does a product title actually affect AI visibility?

More than most merchants expect. In our audits, product title quality is consistently the highest-impact single field on AI readiness scores, often accounting for 30-40 points of variance. It's the first thing AI systems parse when evaluating relevance to a shopping query.

Should I rewrite all my product titles at once?

No. Start with your top 20-30 highest-traffic or highest-margin products. Rewrite those, re-audit, and measure the impact before scaling. A phased approach lets you validate results before committing to a full catalog rewrite.

Will rewriting titles for AI hurt my Google Shopping performance?

Not if structured correctly. The formula, Brand + Product + Use Case + Specs, is compatible with Google Shopping's title guidelines. Brand-first titles with natural-language context often maintain or improve Google Shopping quality scores while significantly boosting AI readiness.

What's the fastest way to audit my current product titles?

Export your product catalog from Shopify and filter for titles under 60 characters, titles with no brand name, and titles that are only a product category plus a color or size variant. Those are your lowest-scoring titles and the highest-priority rewrites.

Want to see what your product titles are scoring right now?
Check Your Store's AI Readiness →