Google Shopping for car parts: getting fitment into the feed.

By Maxxd · Published 21 September 2026

How Google matches a part to a search

Google Shopping matches queries against your product data — mostly the title, then the description, product type, category and attributes — plus what it can read from the landing page. There's no vehicle selector and no compatibility table. If the vehicle isn't named in your data, the part won't appear for a vehicle search, and if a vehicle is named that the part doesn't fit, you'll pay for clicks from the wrong buyers and eventually get disapproved for misleading data.

So the job is to get accurate fitment into the fields Google reads, at scale, from your master fitment data — not by hand.

The fields that matter

1. Title

The single most important field. A title built from structured data beats a hand-written one every time:

Brand + Part type + Position + Vehicle + Generation + Years + Engine
e.g. "Brembo Front Brake Discs Pair – Ford Fiesta ST Mk7 2013–2017 1.6 EcoBoost – 278mm"

Where a part fits many vehicles, lead with the highest-volume one and keep the rest for the description and attributes. Don't cram six vehicles into the title; it truncates and it looks like spam. Google shows roughly the first 70 characters in most placements, so the part type and primary vehicle need to be in there.

2. Description

Put the full fitment list here, in plain text, one vehicle per line: make, model, generation, years, engine. This is where the long tail lives — the second, third and tenth vehicle a part fits. Generated from the master, never typed.

3. product_detail attributes

Merchant Center lets you attach structured name/value pairs to a product using the product_detail attribute (section name, attribute name, attribute value). Use it for fitment and specs: "Fitment / Make / Ford", "Fitment / Model / Fiesta ST Mk7", "Fitment / Years / 2013–2017", "Specification / Disc diameter / 278mm". It's structured, repeatable, and gives Google unambiguous signals to match on and to show as product highlights.

4. Category and product type

google_product_category should be the most specific node under Vehicles & Parts > Vehicle Parts & Accessories — brake discs, not "Vehicle Parts". product_type is your own taxonomy; use it to mirror your site: "Brakes > Discs > Front". Both improve matching and let you build campaign structure around them.

5. Identifiers: brand, GTIN, MPN

Brand is the manufacturer, not your shop. GTIN (barcode) if the manufacturer issues one; MPN always. Google uses these to match your product to its catalogue and to other sellers' listings — get them right and you inherit the product's search relevance; get them wrong and you get disapprovals for "incorrect product identifiers", which is the most common auto-parts problem we see.

6. Custom labels

Five free-text slots (custom_label_0–4) that Google ignores for matching but you can use for campaign control. Label by vehicle make, by margin band, by stock depth, by whether fitment is verified. Then bid up the profitable, in-stock, verified lines and starve the rest.

7. The landing page

Google reads the page the product links to. Product structured data (name, brand, MPN, GTIN, offer, availability) must agree with the feed or you'll get mismatch warnings. And a proper fitment table on the page — the same data, rendered — reinforces what the feed says. See our parts ecommerce page for what that looks like.

Common disapprovals on parts feeds

• Incorrect product identifiers — brand set to the retailer, made-up GTINs, MPN missing.
• Misleading or ambiguous content — a vehicle in the title the part doesn't fit, or "universal" parts with a specific vehicle named.
• Mismatched value (price / availability) — the feed and the landing page disagree because the feed runs daily and prices change hourly.
• Image issues — supplier images with watermarks, text overlays or the part on a busy background.
• Category mismatch — everything filed under a top-level node.

What a good parts feed pipeline looks like

1. Master catalogue and fitment data keyed to K-Type, cleaned and de-duplicated.
2. Rules that generate titles, descriptions and product_detail rows per product from that data.
3. Category and product-type mapping tables.
4. Identifier validation before anything ships.
5. Stock and price pushed frequently (hourly or better); full catalogue daily.
6. Disapproval monitoring with someone whose job it is to act on it.

That's what we build under marketplace feeds, and it's the foundation the paid ads work stands on. If your Merchant Center is a wall of red you've stopped reading, the free fitment audit includes a feed-health check.