Product Data for Pipes & Installation: Small-Part Longtail Meets the Norm

Thousands of near-identical small parts, distinguished only by dimension, thread and norm — where DATANORM and ETIM structure the core, and where AI has to take over the longtail.

Jakob Feinböck, ProductbayJuly 4, 20267 min read
☝️Key takeaways
  • Pipes and installation is a small-part longtail: fittings, connectors and valves where thousands of articles differ only by dimension, thread or norm.
  • One wrong DIN/EN value or dimension makes a part unbuyable — the data value sits in a handful of exact, mandatory attributes.
  • Standards like DATANORM and ETIM structure the branded core, but thin out in the accessory longtail and deliver no sales content.
  • Productbay consolidates every supplier source and uses AI to parse and normalize the norm attributes — with a review step, because here a wrong dimension is a return.

A pipe fitting is one of the least glamorous articles in retail — and one of the most unforgiving. A 15 mm compression elbow and an 18 mm compression elbow look almost the same in a product photo, sit in the same product group, cost roughly the same. But they are not interchangeable, and a customer who orders the wrong one gets a part that does not fit. In the installation assortment, the entire value of a product record collapses into a few exact numbers: dimension, thread, material, pressure rating, norm.

Product data for pipes and installation is a small-part longtail where a handful of exact norm attributes decide whether an article is usable at all. That combination — huge SKU counts of near-identical parts, plus zero tolerance for a wrong value — is what makes this assortment hard. This is a sub-branch of the broader plumbing & heating (SHK) challenge, and it overlaps heavily with the DIY & hardware world, where the same fittings show up on retail shelves.

What makes a small-part longtail like fittings so hard?

The difficulty is not complexity per article — a fitting has maybe eight relevant attributes. The difficulty is volume multiplied by precision:

  • Near-identical articles: thousands of SKUs where two records differ by a single millimeter or a single thread size. The distinguishing attribute is easy to miss and impossible to fake.
  • Zero tolerance for errors: a wrong dimension is not a cosmetic flaw. It produces a return, a frustrated customer and a support ticket. The data has to be exactly right, not approximately right.
  • Fragmented supplier feeds: every wholesaler and manufacturer delivers its own Excel or export, with its own column names — one calls it "Nennweite", the next "DN", the next just puts it in the article title.
  • Longtail depth: the accessory and consumable parts — seals, clamps, adapters, reducers — make up the bulk of the SKU count and almost never arrive with clean, structured data.

Maintained by hand across many suppliers, this simply does not scale. The fix is the standard one — consolidate, normalize, enrich and publish — but here the normalize step carries almost all the weight.

Which norm attributes matter — and where do the standards stop?

The installation trade does have connecting grids. DATANORM is the long-standing exchange format for commercial and basic article data in the plumbing and heating trade, and ETIM classifies articles into technical classes with defined feature sets. Both give the branded core a shared structure. But it is worth being honest about what they cover and what they leave to you:

Data layerWhat DATANORM / ETIM deliverWhere it stops
Article & commercial dataDATANORM carries prices, article numbers, basic dataLittle enriched content, no sales copy
Technical classificationETIM classes with defined feature listsThin coverage for accessory and niche parts
Norm attributes (DN, PN, thread)Present for listed core articlesMissing or in free text for longtail suppliers
Sales content & imagesNot the job of a classificationDescriptions, SEO text, product images absent
Own-brand & niche partsNo standard feed at allEverything arrives as Excel or PDF datasheet

The attributes that actually decide fit — nominal dimension (DN or metric/inch), thread type and size, material, pressure rating (PN), connection type and the governing DIN/EN norm — must be treated as mandatory structured fields. When a standard delivers them cleanly, great. When it does not, they sit in an article title or a PDF, and someone has to extract them. That extraction is the real work.

How does Productbay handle the installation longtail?

The throughline is a three-step job, and in this assortment the enrichment step is where the leverage is — that is exactly what Productbay is built for:

  • Consolidate: import every source once — DATANORM file, ERP export, supplier CSV, Excel, feed URL, FTP, API — and match on GTIN/EAN or article number so existing parts update and new ones are created. The branded core and the accessory longtail land in one catalog.
  • Enrich: AI parses dimension, thread, material, PN and DIN/EN values out of titles and PDF datasheets, assigns ETIM-aligned classes, writes descriptions, translates via DeepL and fills gaps from whitelisted sources — always with a review queue, because a wrong dimension in this assortment is a return, not a typo.
  • Publish: two-way sync to Shopify and Shopware, ERP connections (Xentral, weclapp) and feed exports for Amazon, OTTO and Kaufland — each with per-channel transformations.

Crucially, Productbay starts where DATANORM and ETIM end. If your wholesaler feed already covers the listed core, Productbay complements it and takes over the suppliers outside the standard, the norm attributes buried in datasheets, and the sales content no classification provides. Productbay is built for specialist retailers running multi-supplier, multi-channel catalogs — from mid-sized shops to large chains.

Frequently Asked Questions

Let's look at your product data process

Fittings, pipes and valves by the thousand, each defined by one exact dimension or norm value. See how Productbay consolidates the installation longtail, parses the norm attributes with AI and publishes clean, buyable articles — in a 30-minute walkthrough.

Get started