Today you’ll find (and fix) gaps in titles, descriptions, variants, and identifiers so AI systems can accurately understand and recommend your products.

What you'll learn today

Today you'll achieve these two things:

You'll create an audit template for 10–20 products that you can use right away.
You'll assess title/subtitle, description, variants, and IDs to find what's missing.

Definitions (so this is measurable)

Title: the primary product name shown in search and listings.
Subtitle / key attributes: short qualifiers that prevent ambiguity (material, size range, compatibility, use-case).
Variant: a purchasable option that changes what the customer receives (size, color, pack size).
SKU: internal identifier that should be unique per variant for operations and tracking.
GTIN: standardized product identifier (when available) used across commerce systems.

Why this matters to you

AI prefers clear, unambiguous data. Messy variants lead to wrong recommendations. Without clean data, AI can't accurately quote your products.

Success criteria (metrics)

ID coverage: % of variants with SKU present (and GTIN where applicable).
Variant clarity: variants use one consistent attribute per option (e.g., Size and Color are separate options, not mixed labels).
Title clarity: titles include product type + the minimum distinguishing attributes (so two products don’t look identical).
Description usefulness: first 3–5 lines contain the key specs and decision info customers need.

What to check: product data elements

Title/subtitle

Check that titles are:

Concise with key attributes
Include product type, key features
Clear and descriptive

Description

Check that descriptions are:

Short and focused
Key specs up top
Policy links included

Variants

Check that variants are:

Size/color clear
No mixing of attributes
Each variant clearly labeled

IDs

Check that identifiers are:

SKU filled for each variant
GTIN filled where available
Brand name included

Examples: what works, what doesn't?

✅ Good product data

"Runner Pro, men, blue, GTIN…, SKU…, subtitle: stability, cushioning, shipping 3–5 days."

❌ Poor product data

"Pro shoe" – missing variant info, no ID, no clear attributes.

Common pitfalls (what to avoid)

Using vague titles that omit the distinguishing attribute (customers and AI can’t tell products apart).
Putting key specs only at the bottom of a long description (important details get ignored).
Mixing attributes in variant labels (e.g., “Blue / Size M / Pack of 2” as one value) instead of separate options.
Leaving SKU blank “for now” (breaks tracking and makes duplicates likely).

Practice: create your audit template (10-15 min)

Now you'll create an audit template and review 10 products. Here are the steps:

Create an audit sheet with these columns:
- Product
- Title
- Description
- Variant label
- SKU
- GTIN
- Brand
- Notes
Fill for 10 products.
Compute quick metrics:
- SKU coverage = variants with SKU / total variants
- GTIN coverage (if applicable) = variants with GTIN / total variants
- Count ambiguous titles (could be confused with another product)

Independent practice: log issues (5-10 min)

Now log 5 issues (e.g., missing SKU/GTIN, messy variant) and mark for fix.

Self-check

Before moving on, check that:

✅ Audit template done
✅ 10 products reviewed
✅ 5 issues recorded
✅ You can state your SKU coverage %

If you want to go deeper

Shopify product data: https://help.shopify.com/en/manual/products/products
Google Merchant Center product data: https://support.google.com/merchants/answer/7052112

Shopify product data audit: title, description, variants