ACES and PIES: The Fitment Data Your Wiper Supplier Probably Doesn't Have Yet

The vendor packet arrives, and somewhere on page four it asks for ACES and PIES files. If you are sourcing wiper blades from China, that line is what stalls the program — not tooling, not the strip compound, not the price. ACES and PIES fitment data for wiper blades is what a US retailer or marketplace uses to decide which of your SKUs appears when a shopper picks a year, make and model. A wiper listing without fitment does not sell, and a fitment file that fails validation does not go live.

The uncomfortable part is that your factory almost certainly cannot hand you this data, and the reason is structural rather than lazy. It is worth understanding exactly where the boundary sits, because everything on the factory side of it is cheap to get and everything on your side of it is a project with a budget and an owner.

What the two files actually are

Both standards are published by the Auto Care Association, and both are XML. They do different jobs and are usually built by different people.

ACES — the Aftermarket Catalog Exchange Standard — is the application file. Each record says: this part number, in this position, fits this vehicle. The vehicle is not written as free text. It is written as an ID drawn from the Vehicle Configuration Database (VCdb), typically a BaseVehicleID plus sub-model, engine and other qualifying attributes. The part is classified by a part-type ID from the Product Classification Database (PCdb). Qualifying language that does not fit a structured field goes into a coded qualifier from the Qualifier Database (Qdb).

PIES — the Product Information Exchange Standard — is the product file. It carries the item identity (your part number, GTIN, brand), coded descriptions at several lengths, attributes from the Product Attribute Database (PAdb), packaging segments with dimensions and weights at each packaging level, digital asset references, and interchange or kit relationships.

The practical distinction: PIES describes what the thing is. ACES describes what it goes on. A US channel usually wants both, validated, and stamped against a recent reference-database release.

Why your factory does not have it, and why wipers are the hard case

Four reasons, and none of them are about capability.

The vehicle database is the wrong one. ACES is built against a North American vehicle population. A Guangdong wiper plant builds its cross-reference against the vehicles in front of it — the domestic parc plus whatever export markets it already serves. Zanyu Automotive, for example, names Vietnam as its first stated target export market. A cross-reference tuned to Southeast Asian traffic maps onto US BaseVehicleIDs only in the overlap, and the overlap is not the whole US parc.

The data model is different. Most Chinese wiper catalogues are keyed to blade length and connector type, then cross-referenced to competitors' model numbers — "equivalent to part X" — because that is how the domestic trade buys. ACES does not accept "equivalent to." It wants a vehicle ID. Converting a competitor cross-reference into ACES applications is a licensed, auditable exercise, and doing it by copying another brand's published application table is an intellectual-property problem you do not want.

The brand code belongs to you. ACES and PIES records carry a brand identifier issued to the brand owner. On a private-label program — and Zanyu's stated offer is private-label and custom-specification builds for wholesale and brand buyers, from a stated MOQ of 1,000 pieces — the brand owner is you. The factory cannot supply an identifier it does not hold, and if it did, the data would be filed under someone else's brand.

The barcode belongs to you too. This is where private-label buyers lose weeks. If you let the factory print a barcode on the pack, that GTIN sits under the factory's prefix, or under no prefix at all. Retailers validate GS1 ownership against the brand on the pack. Assign your own GTINs from your own prefix, send them to the factory as artwork data, and keep the assignment sheet as the master.

Wipers are one of the harder categories to map

Buyers who have built ACES for a bolt-on hard part are often surprised by how badly wipers behave.

A single vehicle takes two or three different parts. Driver and passenger blades are usually different lengths, and a rear blade is a third length with a different connector. That means position values — front left, front right, rear — carry real meaning in the file, and a fitment that omits position is worse than useless because the shopper gets the wrong length.

Multi-fit blades explode the record count. A conventional part fits a few hundred vehicles. A multi-fit blade in a single length, with an adapter set covering several arm types, may fit tens of thousands of vehicle configurations. Zanyu's published range includes multi-fit models — the Z816 rubber and Z818 silicone-coated — alongside the Z618 universal U-hook blade and the Z908 hybrid, across universal, multi-fit, hybrid and three-section series. Each of those SKUs, mapped honestly, is an enormous ACES application block, and file size alone will break some receivers' upload limits. Plan for that before you commission the build.

"Universal" fits nowhere in the standard. ACES describes vehicle applications. A blade sold as universal, with fitment determined by the adapter the customer selects, is not really an application record — it is a length plus a connector list. Some channels want it published with full fitment anyway; some want it flagged as a non-application item and merchandised by size. Ask your receiver which, in writing, before anyone builds anything, because the two answers produce completely different files.

Length tolerance is a returns problem, not a data problem — until it becomes both. If your PIES attributes say 24 in and the ACES application says the vehicle takes 600 mm, you have published a contradiction that a customer will find. Fix the attribute source of truth first.

Who builds the data, and what you are actually paying for

Three routes, in ascending order of control and cost.

1. A catalogue service bureau. Specialist firms build ACES and PIES for aftermarket brands. You send a part list with attributes; they return validated files and often handle submission. Fastest route for a first program.
2. In-house. You subscribe to the reference databases, license or buy catalogue software, and hire or assign a catalogue manager. Only worth it past a few hundred active SKUs.
3. Through the receiver. Some channels accept a spreadsheet fitment template as an on-ramp for new vendors and convert it themselves. This is the cheapest first step and worth asking about explicitly — it is rarely advertised.

Whichever route you pick, insist that the quote is broken out by line so you can compare them. The cost lines are:

• Association membership and reference-database subscription, annual
• Initial ACES build, usually quoted per part number and sensitive to how many applications each part carries
• PIES attribute population, per SKU
• Digital asset production — photography to the receiver's spec, per SKU, almost always separate
• Data-pool or receiver submission and hosting, annual
• Maintenance: re-stamping against each reference-database release, extending applications to new model years, and re-issuing on any pack change

That maintenance line is the one first-time buyers forget. Fitment data is not a deliverable, it is a subscription. Receivers reject files stamped against a stale reference version, and new model years arrive whether or not your catalogue budget did.

One more cost that is not on any invoice: images. Factory studio photographs almost never pass a US retailer's image spec — minimum pixel dimensions, pure white background, no watermark, no packaging overlay, defined file naming. Budget for reshooting the range in-market, and treat any supplier-supplied image as a placeholder for internal use.

What you can get from the factory, and how to ask for it

Everything in the PIES item and package segments is the factory's to give, and most of it is a five-minute job for them if you send a filled-in template instead of a request. What is not published for this supplier is the per-model attribute sheet: the exact blade lengths offered per series, the adapter list, what ships inside the pack, piece and pack weights, and carton dimensions. Ask, and ask in the format you need back.

Send a spreadsheet with one row per SKU and these columns, pre-filled with the units you want:

A supplier who fills that sheet accurately in a couple of days is telling you something useful about how their documentation is run. A supplier who returns it with blanks or with round numbers that do not reconcile is telling you something too. Run the same sheet against the sample you receive and measure the pack yourself; sampling is stated at 3–7 days and volume production at 15–20 days, so there is time to check before the artwork locks. The same discipline that goes into a written product spec sheet for a Chinese factory applies here — the template is the specification.

Common questions

Can I just use the factory's cross-reference table as my fitment?

No, for two reasons. It is usually keyed to another brand's part numbers rather than to vehicle IDs, and republishing another brand's application table carries an intellectual-property exposure that your channel's vendor agreement will make your problem. Use it as a research input for whoever builds your ACES file, not as the file.

Who owns the finished ACES and PIES files?

Whoever is named as the brand. Get ownership and portability written into the service-bureau contract before the build starts, including the right to take the source data with you in a documented format if you change providers. Brands that skipped this have paid to rebuild a catalogue they already owned.

How does this interact with private-label branding?

Directly. Under a private-label or white-label arrangement, you are the brand of record for the channel, the data, the barcode and the origin mark on the pack. The factory manufactures; the catalogue is yours. The only parts the factory needs to understand are the ones that touch the product: the attribute sheet, the barcode you assign, the pack contents, and the fact that a pack or carton change forces a data re-issue. Write that last point into the purchase order — an unannounced carton downsize is a data incident, not just a freight surprise. And allow more calendar than the production lead time: catalogue build is a parallel workstream that starts when you approve the sample, not when the container books.

What to ask the supplier next

Before you commission any catalogue work, send the factory this list and see what comes back:

1. A filled attribute sheet, one row per SKU, in the columns above, with dimensions measured rather than nominal.
2. Any application, cross-reference or fitment table you hold, in any format, stating what vehicle population it was built against.
3. Confirmation that the pack carries no barcode unless you supply the number, and that your assigned GTINs will be printed exactly as issued.
4. A statement of which pack and carton formats are available at the stated MOQ of 1,000 pieces, and what a format change does to the lead time.
5. Written notice terms for any change to pack or carton dimensions, since that change forces a data re-issue.
6. Sample packs of each format, so you can measure and photograph rather than take a drawing on trust.

Zanyu Automotive is a workable example of how this plays out with a narrow-range wiper maker: four published series, model numbers on the public profile, private-label and custom-specification programs offered, and no published fitment data of any kind — which is the normal starting point, not a red flag. The Zanyu Automotive factory profile on CMH sets out what has been checked and what has not, and the same MOQ and lead-time discipline that applies to any China sourcing data set applies to the attribute sheet you are about to send. Get the attributes first. Budget the fitment separately. Do not let a listing date depend on a file nobody has been asked to build.