What Auto Parts Brands Get Wrong With ACES and PIES (and How to Fix Each One)

Most auto parts brands do not have a product problem. They have a data problem they cannot see. The part is good. The catalog data that tells a retailer's system what the part is and what it fits is quietly wrong, and that is what costs the sale. Below are the mistakes we find in almost every catalog we touch, and the fix for each one. If any of these sound like your files, the good news is that all of them are fixable without deleting a single real application.

1. Fitment buried in free-text notes

The single most common and most expensive mistake. A brand knows its parts distinguish on tank size, spline count, or duty level, so someone writes that distinction into a free-text note. A human can read it. No validator, lookup, or marketplace can. To every system downstream, two different parts now look identical on the same vehicle.

Why it costs you: free-text notes cannot be filtered, do not travel between systems, and are the raw material of overlaps. A shopper sees two parts, cannot tell them apart, and either picks wrong (a return) or leaves (a lost sale).

The fix: convert every note that carries fitment into a coded qualifier. Sizing notes become parameterized qualifiers. The distinction that lived in prose becomes something a system can actually act on. This is the core of our Zero-Loss Cleanup method, and it is where most catalogs go from failing to passing.

2. Overlaps nobody is measuring

An overlap is two parts mapped to the same vehicle with nothing coded to separate them. Left alone, they multiply into the hundreds of thousands on a mid-size catalog. Brands often do not know how many they have because nothing is counting.

Why it costs you: overlaps are the number one reason a file fails an industry assessment, and the number one reason a lookup cannot decide which part to show.

The fix: measure the overlap count, then drive it down with coded qualifiers, complementary qualifiers on the broad part, and tier mapping, measuring again after every pass. On one performance driveline brand we took overlaps from 301,762 to zero without deleting a single application. The measuring is the discipline. Never batch changes blind.

3. Deleting coverage to fake a clean number

The shortcut that looks like a fix and is not. When overlaps are hard to resolve, it is tempting to just delete applications until the number looks clean. The file passes. The brand also just made its parts stop showing up on real vehicles.

Why it costs you: every deleted application is a vehicle your part no longer sells to. You traded real revenue for a green checkmark.

The fix: qualify, never delete. Coverage before must equal coverage after, verified. If a distinction is real, code it. If two parts truly overlap, that is a decision for the brand, not a silent deletion.

4. Template PIES descriptions

Fifteen thousand SKUs, one paragraph of copy with the part number swapped in. It is fast to produce and it quietly kills search performance.

Why it costs you: duplicate content does not rank, does not convert, and tells a shopper nothing about why this part fits their truck. Retailers increasingly reject or bury template copy.

The fix: vehicle-aware descriptions built from the part's real attributes. On one brand we turned 85 template paragraphs into 4,726 distinct descriptions, each one specific to what it actually fit.

5. Specs trapped in PDFs

The attributes that let a shopper filter (bore, length, thread, amperage) live in a spec sheet or a PDF instead of coded PAdb attributes in the PIES file.

Why it costs you: a shopper filtering for the exact spec they need never sees your part, because the spec is not in a field anything can filter on.

The fix: extract the specs into coded product attributes. On one brand that meant loading 104,297 spec attributes out of PDFs and into the file where they belong.

6. Good-Better-Best labels used as decoration

Manufacturer labels exist to mark a true good, better, best tier on the same vehicle. Brands often slap them on single parts as a marketing flourish, which breaks the assessment.

Why it costs you: misused labels are a scored assessment failure, and they confuse the lookup logic they are supposed to help.

The fix: use a label only when there are genuinely multiple tiers for the same fitment, mapped to real duty qualifiers. Otherwise leave it off.

7. Treating ACES 5.0 and PIES 8.0 as optional

The standards moved. Element renames, new required fields, new packaging segments. Files built to the old shape start bouncing, and the assessment tooling is still catching up to the new versions, which makes the timing genuinely tricky.

Why it costs you: a file in the wrong version format fails at the structure check before anyone even reads your data.

The fix: migrate deliberately, sequence the changes, and keep a change log. This is exactly the kind of transition worth having done for you rather than discovering the hard way on a rejected submission.

The pattern behind all seven

Every one of these is the same underlying issue: information that a person can understand but a system cannot. The fix is always to move that information into a coded, standard field, measure the result, and never trade away real coverage to get a clean number. That is the whole job.

Want to know which of these your file has?

Send us your current ACES or PIES export. We grade it against The Catalog Scorecard and hand you a ranked list of exactly what to fix, in 24 to 48 hours, at no cost. No cleanup first, no sales call. Get your free file audit.

FAQ

Why do ACES files get rejected?

Most rejections trace to structure errors (wrong version format or element order), invalid or missing qualifier values, or overlap and note problems flagged by the receiver's assessment. The first bounce is usually structural; the deeper issues are fitment buried in notes and unmeasured overlaps.

What is the difference between ACES and PIES?

ACES describes what a part fits (vehicle fitment). PIES describes what the part is (product information, attributes, images, packaging). A complete catalog needs both, and they must agree on which parts exist.

Can you fix ACES and PIES data without losing coverage?

Yes. The correct method qualifies parts rather than deleting them, so the same vehicles are covered before and after. Coverage should be verified identical, not assumed.

How long does it take to clean an ACES file?

A graded audit takes 24 to 48 hours. A full cleanup depends on catalog size, but the method scales: we have cleaned catalogs from a few thousand applications up to over half a million.