Missing value: color / size / gender / age group

What Google requires

Four attributes on every apparel product

In the major apparel markets, every clothing product needs color, size, gender and age_group. Gender and age group are closed lists (male, female, unisex; newborn through adult). Color and size are free text with rules of their own: a real colour name rather than a hex code or article number, one value rather than a concatenation. Variants of one garment share an item_group_id and differ on at least one of these attributes.

What happens if you do not

Per-attribute disapprovals across the whole clothing range

Each missing attribute produces its own disapproval, so one structural gap in the export multiplies across products and attributes into hundreds of flags. The subtler failures come from values that exist but do not qualify: a colour written as #1F2A44, a size and colour concatenated into one field by the shop system, or an entire group of variants sharing one product page URL so that Google cannot tell them apart.

How to fix it by hand

Fix the mapping, then the values

• Map the four attributes from the shop's variant data into the feed; the data almost always exists, unmapped.
• Write colours as names, split concatenated values into their fields, and keep one value per attribute.
• Give every variant its own row with a shared item_group_id, and check that the group actually varies on colour or size.

Which of our checks apply

12 checks apply, read from the scanner's registry

The apparel checks fire per attribute and per product, and the colour logic repairs what it can prove: a hex code becomes a name only when your own data contains that name.

• color_missing_apparelat risk
• size_missing_apparelat risk
• gender_missing_apparelat risk
• age_group_missing_apparelat risk
• color_codeblocks
• color_concatenatedblocks
• color_not_a_colourblocks
• size_separatorat risk
• gender_valueblocks
• age_group_valueblocks
• item_group_id_missingat risk
• item_group_no_variant_attributeat risk

The quicker way

Let the scan find them

One scan of an apparel feed shows every product-and-attribute pair Google will flag, before the per-attribute disapprovals multiply. The scan to identify errors is free and shows every finding before you pay anything. You only pay for the fix.

Scan your feed