Customer master hierarchies for OEMs: linking end customers, dealers and distributors
Your ERP knows who you invoiced. It rarely knows who owns the machine. How to model end customers, sites, dealers and distributors so installed-base and aftermarket analytics add up.
Not ready to talk yet? Take the 3-minute Data Foundation Check
TL;DR
- The sold-to party is often a dealer. The machine sits at an end customer’s site. Most aftermarket analytics break right there.
- Model three things separately: legal hierarchy (group → company → site), channel (who sold or serves it) and machine location.
- Deduplicate accounts first. The same customer often exists under several numbers per country and company code.
- Without this, wallet share and leakage are calculated on the wrong account.
Why does the customer master break aftermarket analytics?
Machines are sold through dealers, to EPC contractors or to a holding company. Parts are bought later by the plant, maybe through a different distributor. If each of those is an unrelated record, the installed base looks scattered and parts revenue can’t be matched to machines.
| Role | Example | Where it usually sits |
|---|---|---|
| Group / parent | Food group with 12 plants | Often missing, or only in CRM |
| Legal entity | Plant operating company in Poland | Customer master (in S/4HANA: Business Partner) |
| Site | Plant address where the machine runs | Ship-to partner |
| Dealer / distributor | Regional dealer who sold the machine | Sold-to partner on the machine order |
| Service payer | Whoever pays for service | Payer / bill-to partner |
What should the target model look like?
- Legal hierarchy group → company → site, maintained once.
- Machines link to the site, not to the dealer.
- Dealers and distributors link to machines and sites as a relationship with a role and a validity period.
- Orders keep their partner functions (sold-to, ship-to, payer), mapped onto the model.
Keep channel and ownership apart
A dealer who sold 40 machines in 2015 doesn’t own them. Mixing the two makes dealers look like large plants, and the real end customers disappear from the analysis.
How do you build it? 5 steps
- Deduplicate accounts. Match on normalised name, address, VAT ID and D-U-N-S number where available.
- Build the legal hierarchy. External data such as commercial registers or D-U-N-S family trees helps for top accounts. The long tail is often fine as a flat list.
- Assign machines to sites. Start from ship-to addresses on machine orders, then correct with service visit addresses, which are usually more current.
- Model channel relationships. Dealer → machine and dealer → site, each with a role and from/to dates.
- Set rules for new accounts. A duplicate check and a required parent at creation stop the problem from coming back.
A simple account match key
MATCH KEY = normalise(name) + postcode + house number normalise(name): lowercase, unify umlauts (ü → ue) remove legal forms (GmbH, AG, KG, Sp. z o.o., S.r.l., Ltd) remove punctuation, unify & / and / und "Müller Lebensmittel GmbH & Co. KG" → mueller lebensmittel "MUELLER LEBENSMITTEL GMBH" → mueller lebensmittel
What does it unlock?
| Analysis | What it needs |
|---|---|
| Wallet share per customer group | Group hierarchy + machines per site |
| Leakage by channel | Dealer relationships + parts orders per site |
| Key account pricing | Group hierarchy across countries |
| Retrofit targeting | Machines per site with age and type |
| Service contract coverage | Machines per site + contract per machine |
Checklist
- Duplicate check run on the customer master, across company codes
- VAT IDs captured for EU business accounts
- Machines linked to a site, not only to a sold-to party
- Dealer relationships stored with role and dates
- Required parent and duplicate check for new accounts
FAQ
What is a customer hierarchy?
A structure that links related customer records: a group or parent company, its legal entities and their sites. It shows total business per group across countries and company codes.
Should machines be linked to the dealer or the end customer?
To the end customer’s site. The dealer is linked as a relationship with a role, such as seller or service partner, so both views stay available.
How do we find duplicate customer accounts?
Normalise names by removing legal forms and unifying spelling, then compare with address, VAT ID and identifiers such as D-U-N-S. Review uncertain matches manually.
Should the hierarchy live in ERP or CRM?
Wherever it is maintained reliably, as long as other systems use the same IDs. Many companies keep the legal hierarchy in ERP or MDM and the sales view in CRM.
Does this matter for S/4HANA?
Yes. In S/4HANA customers and suppliers become Business Partners. Cleaning duplicates and hierarchies before migration avoids carrying the same problems into the new system.
Do you know who owns your machines?
Send us customer master and machine sales exports under NDA. We’ll show you how many accounts collapse into one and where your installed base really sits.





