Master data management consulting for companies in Germany
Your customers and products live in several systems, each with its own number and its own copy of the address. I define with you what counts as one customer or product and which system leads for each of its fields. The other systems then follow that record. I take on freelance and project work, remote from Frankfurt. Let's talk about your needs in a free 30-minute video call.
What I take care of for you
-
Experience across systems
I have designed data models for more than twenty years and connected a customer portal to an existing CRM system. Your staff should find one customer where today they may find three.
-
One record per customer
We write down what counts as one customer, supplier or product in your company. Each system then points to that record, whatever number it gave it.
-
Keys your systems already share
Commercial register numbers, VAT IDs and article numbers find matching records across systems. A contradiction between two records stops the match, and unclear pairs go to a person.
-
Links that can be withdrawn
Two records of one company get a link, and both keep their history. A wrong link comes off in one step, and no data is lost.
-
Changes through one door
Other systems write through an API, each with its own access token. An import may add records and cannot overwrite what your staff maintain.
When one customer lives in four systems
Master data are the records that other records point to: customers, suppliers, products and sites. Master data management decides what counts as one of them and which system holds the leading version of each field.
In a growing company these records tend to appear system by system. The CRM knows a customer under one number, the ERP under another, the shop under an email address and accounting under a debtor number. Each system keeps its own copy of the address, and each copy changes on its own. A simple question then has several answers, such as how many active customers the company has.
What I set up
A record describes one thing in the world, and its type says what kind of thing it is: a company, a person, a product or a site. The number a system gives the record is a handle for it and says nothing on its own. The model behind this is on Tracking objects over application boundaries.
The commercial register number, the VAT ID and the article number exist outside your systems, and several of your systems hold them. I use such keys to find records that describe the same thing. Registers reuse codes, and two records with one code can still differ in a basic fact, so the key proposes a match and a check decides it.
When two records describe one company, a link says so, and both records keep their numbers and their history. Withdrawing a wrong link is one change. A contradiction, such as a dissolved company and an active one, blocks the link.
The role of a contact person at a customer is a record of its own, and so is a contract between two companies. A contact can leave while both companies continue, and the end of the role is stored as a change with its date.
The other systems write through an API, each with an access token of its own. An import token may create records and may not change existing ones, so a faulty import cannot overwrite what your staff maintain. How other systems couple to the store describes these rules as I apply them in client work.
Who owns which field: an example
A field table names the leading system for each field of a record. The example below is for a customer. Yours will differ, and we write it together in the project.
| Field | Leading system | Who changes it | How the other systems check it |
|---|---|---|---|
| Legal name and legal form | ERP | Accounting, from the commercial register | Commercial register number |
| VAT ID | ERP | Accounting | Format check and the EU’s VIES service |
| Billing address | ERP | Accounting | Postcode fits the country |
| Delivery addresses | ERP | Order processing | One delivery address per site |
| Contact persons and their roles | CRM | Sales | Email domain fits the company |
| Newsletter consent | Shop or CRM | The customer | Date and source of the consent |
| Customer number | ERP | Assigned once, not changed | Stored in each system as a reference |
For a product, the same table names the ERP for the article number and the price, and the shop or a product information system for texts and pictures.
Who I am
I am Michael Wutzke, based in Frankfurt, with more than twenty years in IT and media. At Policen Direkt I connected a customer portal for insurance policies to an existing CRM system. I have helped build a client’s information platform on which companies and people keep one identity across registers and sources. I hold a training authorization from IHK Frankfurt and teach at Claude Hacker House. I build software with Claude and Codex and am interested in open-source AI models that a company runs on its own servers. Details: Career stages.
How the project runs
-
Free video call
In 30 minutes we talk about the systems that hold your customers and products and where they disagree.
-
Assessment together
We list the systems, the fields they share and the people who change them.
-
Quote and order
Based on the assessment you receive a quote. The work begins with your order.
-
Definitions and field table
Together we write down what counts as one record per type and which system leads for each field.
-
Matching
Shared keys and names find records that describe the same customer. Contradictions block a match, and unclear pairs go to your staff.
-
Connections
The leading system passes its changes to the others through their APIs, and each write carries its source.
-
Handover
Your team receives the definitions, the field table and the matching rules, and extends them when a new system arrives.
Questions companies ask
Do we need master data management software?
Not to start. The definitions and the field table come first, and they show whether a dedicated tool pays off. One of your existing systems can take the lead, for example the ERP for customer numbers.
What happens when two systems disagree about a customer?
The field table names the leading system for each field, and its value wins. The other value stays in the history with its source, so your staff can see what the other system said.
Why not use the customer number as the shared key?
A customer number belongs to one system. A VAT ID or a commercial register number exists outside your systems, and I store such codes as text, because a leading zero is part of the code. Registers can reuse a code, so a shared key suggests a match and a check decides it.
Which systems can you connect?
Systems with an API or a database I may read: CRM, ERP, shop, accounting and your own applications. I build the connections with REST APIs and scripts in PHP, Python or Node.js.
How does this differ from data quality work?
Master data management settles which record leads and how the systems share it. Data quality work cleans the records inside a system. Details: Data quality consulting.
Which engagements do you take on?
Freelance and project work, remote first. I live in Frankfurt and come on site for the workshops on definitions, where the people who know the records sit at one table.
Details on the work behind this page
-
Tracking objects over application boundaries
Many records describe the same thing, and systems rarely agree when two of them mean one object. I want to find out how to decide that.
-
Maintaining data quality is a challenge
AI agents read doubtful sources and write to shared data. Can one rule set keep it reliable?
-
A digital insurance manager on AWS
An insurance manager in Angular, life insurance portals and AWS infrastructure in two regions.
Your master data management consultant in Germany
I am Michael Wutzke, a master data management consultant in Germany, based in Frankfurt. In a free 30-minute video call we look at the systems that hold your customers and products, and you learn which of them could lead and what a first field table would cover.
Searches this page answers
- master data management consulting
- master data management consultant
- master data management
- master data management (mdm)
- what is master data management
- what is master data management in simple terms
- what is master data
- master data management meaning
- master data management example
- master data management best practices
- customer master data management best practices
- product master data management best practices
- supplier master data management best practices
- master data management process
- master data management process steps
- customer master data management process
- master data management strategy
- mdm implementation
- mdm implementation steps
- master data management implementation steps
- master data governance
- master data governance framework
- master data vs transactional data
- master data vs transactional data examples
- customer master data
- customer master data management
- product master data
- product master data management
- supplier master data management
- vendor master data management
- master data quality
- master data cleansing
- master data management erp
- master data management vs crm
- entity resolution
- duplicate customer records
- single customer view