Guide
What Is Master Data Quality?
Master data quality is how well your core records, such as customers, products and price lists, are complete, valid, unique, consistent, timely and accurate. It matters to finance because every invoice line inherits its customer, price and payment term from master data. One customer stored under two codes is enough to bill the wrong price and split discounts and terms.
Last updated:
By the korppi team, Cemal Tosun and Emin Genisol
Master Data vs Transactional Data
Master data describes the stable things a business works with: customers, products, suppliers, price lists, payment terms and sales reps. Transactional data records events, such as orders, invoices, credit notes and payments. A single customer record can sit behind thousands of invoice lines, so one error in master data is repeated on every one of them.
That is why fixing invoices one by one rarely works. If the customer still points to an old price list, the next invoice repeats the mistake. Quality has to be fixed and owned at the master record, which is the job of data governance.
67%
of organizations do not fully trust the data they use for decision-making.
The Six Data Quality Dimensions, With Customer Master Examples
Most frameworks, including DAMA-DMBOK, describe data quality through six dimensions. Completeness asks whether the value is there, validity whether it is allowed, uniqueness whether each real customer is stored once, consistency whether related records agree, timeliness whether it is up to date and accuracy whether it matches reality.
On a customer master each dimension has a price. A missing payment term delays cash, a duplicate customer splits discounts, and a price list assigned late bills last year's price. The dimensions are easy to name. Knowing which failures in your data actually cost money, and which are harmless noise, is the part that takes experience.
Why Do Duplicate Customer Records Happen?
Duplicates are rarely one big mistake. They build up from ordinary work: a branch creates a customer again because it could not find it, the same company is spelled three different ways, a new delivery address becomes a new account, or two customer lists are kept after an acquisition. Nobody does anything wrong on the day, and the cost appears months later.
The same mechanism drives late price increases and below-contract billing, described in pricing and discount leakage.
How to Measure Master Data Quality
A quality score on its own tells a CFO very little. What matters is the money behind it: which invoice lines were billed against a broken record, and who owns the fix. Getting from a customer master to that answer means understanding how your ERP links customers, prices and terms, and that is rarely obvious from the data alone.
This is exactly what the Data Health Report does for you on your own data, so it is a good place to start a conversation with us.
Customer Master Data Cleansing, Step by Step
Cleansing looks simple until you try it. Two records that look alike may be different legal entities, two that look different may be the same customer, and a careless merge can break invoice history, open balances and audit trails. Every ERP, including SAP, Logo and Netsis, handles merging and blocking differently, and the right answer depends on your contracts and your reporting.
That is why cleansing works best with expertise and a named owner who approves each decision, rather than a one-off script. We are happy to walk through what that looks like for your system. See also ERP data quality.
Validating Tax IDs: The Turkish VKN and TCKN Example
In Türkiye, legal entities are identified by a VKN (Vergi Kimlik Numarası), and individuals, including sole traders, use their TCKN (T.C. Kimlik Numarası) as their tax ID. A wrong or missing number is one of the most common reasons a customer ends up stored twice.
A number that looks right is not proof of a real, active taxpayer. Whether the taxpayer is registered for e-Fatura also decides whether an invoice goes out as an e-Fatura or an e-Arşiv Fatura, and the wrong choice means cancelled invoices and delayed cash. Keeping that status correct belongs in the customer master too.
How korppi Measures and Fixes Master Data Quality
korppi starts with the data before it reports any money. Duplicate customers, invalid tax IDs and broken codes are brought to the surface and routed to their owner first, so leaks are neither double counted nor missed. Every figure is calculated from your own lines and linked to its invoice lines, and AI explains the results without inventing numbers.
The korppi Data Health Report puts a value on every problem record and shows the before and after. korppi Revenue Intelligence then keeps it current with every new export. A read-only export is enough, korppi never writes to your system, and we agree the details with your team. The security brief is shared on request.
Start With a Data Health Report
The quickest way to know what your master data is costing you is to look at it. Request a Data Health Report for your own data, or request a demo to see korppi first.
Frequently Asked Questions
What is master data quality?
Master data quality is the degree to which core records such as customers, products and price lists are complete, valid, unique, consistent, timely and accurate. Because transactions inherit values from master data, its quality decides whether invoices carry the right price, discount and payment term.
How do you find duplicate customer records?
Some are obvious, many are not. The same company can be spelled many ways, and records that look alike can be different entities. Finding them reliably, and knowing which ones cost money, takes experience with your ERP. A Data Health Report shows them for your own data, with the invoice lines behind each one.
What are the six dimensions of data quality?
The commonly used set is completeness, validity, uniqueness, consistency, timeliness and accuracy. Some frameworks add integrity or conformity. For finance, each one matters only as far as it changes what lands on the invoice.
How do you measure master data quality?
The measure that matters to finance is money: how much revenue is billed against broken records, and who owns the fix. A Data Health Report on a read-only export of your own data shows exactly that.
Should duplicate customers be deleted or merged?
Records with posted transactions should not simply be deleted, because history has to stay linked for audit and reporting. The right way to retire a duplicate depends on your ERP and your contracts, so it is worth agreeing it with the data owner first.