How HVAC distributors manage product data for replacement parts

HVAC distributors manage replacement-parts data by tying every part to three things: the OEM part numbers it replaces, the equipment model numbers it fits, and the nameplate specs a contractor would otherwise match by hand (voltage, horsepower, RPM, rotation, frame and dimensions, refrigerant). The ones who do it well load manufacturer cross-reference and supersession data into the ERP or PIM as structured relationships rather than description text, so a search on a model number, an old OEM number, or a spec combination all land on the same orderable SKU.

The contractor at the unit, and what they actually have

Replacement parts are ordered from the field. The tech is standing at a condenser with a dead fan motor and three possible starting points:

• The equipment model and serial number off the rating plate. Carrier, for example, puts the model and serial on the rating plate, near the service valves on outdoor units and inside the front door panel on furnaces.
• The OEM part number printed on the failed component, if it is still legible.
• The component's own nameplate: HP, voltage, RPM, rotation, frame.

Your catalog has to answer from any of the three. If it only answers from your own SKU or the aftermarket brand's stock number, the contractor calls the counter, or orders from whoever's search does work.

What product data HVAC contractors need when ordering parts online

For a motor, the spec list is long and specific. A motor cross-referencing guide from HVAC Know It All lists horsepower, voltage, RPM per speed, phase, service factor, direction of rotation, plus frame size, mounting type (belly band, stud, resilient and so on), enclosure, single or double shaft, bearing type, motor type, and application. Each is a field a contractor might filter on, and a reason a "close enough" motor comes back.

The rules also differ by motor type, which is why these have to be structured fields and not prose. One troubleshooting reference states that for PSC motors you match horsepower, voltage, RPM, frame diameter, rotation and the run capacitor µF rating, because a PSC motor's speed is fixed to line frequency. Some ECM replacements follow a narrower manufacturer rule: Genteq's Evergreen VS brochure tells techs to match horsepower and voltage, and says RPM and FLA do not need to match when replacing the Genteq motors it covers. A catalog that stores RPM as a hard filter on that kind of replacement will hide valid parts. One that ignores RPM on PSC motors will show wrong ones.

A workable field map for the most-searched repair categories:

Part type	Fields contractors filter or match on
Condenser fan / blower motor	hp, voltage, rpm, phase, rotation, frame_diameter, shaft_count, shaft_diameter, mounting_type, motor_type, capacitor_uf
Run capacitor	capacitance_uf, voltage_rating, shape, dimensions
Contactor	poles, coil_voltage, amp_rating
Compressor, TXV, coil	refrigerant, capacity, voltage, phase, connection sizes
Any OEM-sourced part	oem_part_number, fits_models, superseded_by, replaces

The last row is the one that tends to be missing.

Where HVAC replacement-parts cross-reference data comes from

There is no single cross-reference file. Distributors assemble it from several sources, each with a different shape:

• Aftermarket manufacturers' replacement guides. Replacement-motor and capacitor makers publish which OEM numbers or OEM motor families their stock numbers replace. The Evergreen VS brochure, for instance, says the product line is designed to replace Genteq models 2.3, Eon and 3.0. That is family-level fit, not a per-model list, and should be stored that way.
• OEM parts lists and bills of material. These map an equipment model number to the factory part numbers inside it. Some services sell this as a dataset; Bluon pitches its distributor search as a way to find the original BOM and compatible cross references for any brand.
• Parts specialists' cross-reference lookups. Global Parts Supply's lookup accepts current numbers, OEM numbers, and discontinued or superseded numbers, and its product pages show the OEM number, what the part replaces, what replaced it, and which brands it fits. It also tells buyers to check their number against that fitment list before ordering, which is the right instinct for any cross-reference.
• Your own ERP history. Customer part numbers, counter substitutions, and "wrong part" returns are cross-reference data nobody has structured.

The common problem is format. Distributor Data Solutions puts it plainly on its HVACR page: product data from manufacturers rarely arrives in consistent formats.

How to structure it so search by model number or OEM part number works

Treat compatibility as relationships between records, not attributes on one record.

1. Keep one row per orderable SKU with its typed spec fields (above), its GTIN where the manufacturer assigns one (GS1 US describes a GTIN as a GS1 identification key used to identify a trade item), and its manufacturer part number.
2. Store cross-references as their own table: your_sku, reference_number, reference_brand, relationship_type (oem_equivalent, universal_replacement, supersedes, superseded_by), source, source_date. Distribution ERPs such as 10X ERP advertise the ability to map manufacturer, vendor and customer part numbers to an item; the question is whether you populate the relationship type and source.
3. Resolve supersession chains. If OEM number A was replaced by B and B by C, a search for A must return the current orderable item, and the page should say so. Store each hop; compute the chain at index time.
4. Store model fitment separately from part identity. fits_models is a list of equipment model numbers or model-number patterns. Index it so a pasted model number returns every part with a fitment row for it.
5. Normalize numbers before indexing. Strip dashes, spaces and leading zeros into a search key while keeping the printed form for display, so 12-34567-01 and 123456701 hit the same row.

For illustration: a 15,000-SKU repair-parts catalog averaging four cross-reference rows per SKU is 60,000 relationship rows, each needing a source and a type. That volume is why the work is usually the bottleneck, not the software. If you are weighing systems for this, our comparison of PIMs for HVAC distributors looks at how each handles relationship data.

Search filters for voltage, horsepower, dimensions and refrigerant

Faceted filters only work on clean, single-unit values. 1/4 HP, .25HP and 0.25 hp must normalize to one value, or the horsepower facet shows three buckets. Voltage needs a rule for dual-rated parts (208-230V should match a filter of 230V). Rotation needs a value for reversible motors, not a blank.

Refrigerant deserves its own handling because the installed base is now mixed. EPA's HFC phasedown FAQ says new components may continue to be manufactured and imported so existing systems can be maintained for their useful life. So you will carry R-410A service parts alongside A2L-refrigerant equipment parts for years. Tag refrigerant on parts where it matters (compressors, metering devices, coils, condensing units) and leave it off where it does not (a fan motor). Our post on A2L transition product data covers the field changes in more detail.

Where it goes wrong

• Cross-references in the description field. "Replaces Carrier and Rheem motors" in prose is invisible to a part-number search.
• No relationship type. An exact OEM equivalent and a universal replacement that needs a capacitor change are different promises. Flag the difference.
• Stale supersessions. OEM numbers change; a chain that stops one hop early sends contractors to a discontinued item.
• Specs copied from the wrong source. OEMs warn about this from their side; Carrier cautions that third-party parts carry risks such as mismatched specs. Every spec value should trace to a document.

Who does the work

The data model is easy to design. Filling it is not: someone has to read the replacement guides, extract each OEM-to-SKU row with its type and source, normalize HP and voltage, resolve supersessions, and keep doing it as manufacturers revise their lists.

That is the work Anglera does. Your PIM or ERP stores the data; Anglera extracts values and cross-reference rows from the actual spec sheets, catalogs and manufacturer pages, scores each value's confidence, and flags conflicts (two sources disagreeing on RPM, for instance) for review instead of guessing. It works alongside whatever PIM, ERP or syndication tool you already run, can start from a flat ERP export, and is typically live in about 30 days. Here is how it works end to end, and what it looks like for HVACR distributors specifically.

Deciding where to start

Start with the categories that generate the most "wrong part" returns in your own data. Motors and capacitors are a natural first pick because their match rules are the most spec-heavy. Build the cross-reference table and normalized spec fields for those first. Measure whether model-number and OEM-number searches return an in-stock SKU. Then extend category by category. The HVACR distributor guide covers the rest of the catalog.

A replacement-parts catalog earns contractor orders when the number they are holding finds the part they need. Anglera keeps the sourced specs and cross-references behind that search filled in and current, on top of the systems you already have.