Picture a mid-sized wholesale distributor. Hundreds of customer records have been entered into the system over the past two years. But the same customer appears under three different names: once as ‘Ahmet Tekstil’, once as ‘A. Tekstil Ltd.’, and once simply as ‘Ahmet Bey’. Each entry has its own open balance. By month-end, the receivables report is a mess and nobody can agree on how much the customer actually owes. The software is not the problem. The problem is that nobody set rules for how data gets entered in the first place.
ERP (enterprise resource planning) software keeps all customer, product and supplier information in a single shared pool. That pool is called master data. Customer cards, product codes, supplier addresses — all of it is master data. No matter how capable the software is, if the underlying records contain errors or gaps, every report that comes out of the system will be wrong. Accountants have a blunt way of saying this: garbage in, garbage out. The rule is simple. Applying it consistently in a busy office is not.
In most small and medium-sized businesses, data entry works on an informal basis: whoever learned the system does the entering. When a new customer comes in, sometimes the sales rep opens the card, sometimes the accountant, sometimes the warehouse clerk. Each person follows their own habits. One writes the city, another skips it. One adds the tax number, another leaves it blank. Over time the record pool grows bloated and unreliable. The only way to stop this from happening is to assign a clear owner to each type of data — one person or role responsible for opening, updating and closing each category of record.
In a master data ownership model, every record type has a designated owner. Customer cards belong to the sales supervisor. Product records belong to the warehouse or production team. Supplier cards belong to purchasing. When a new record needs to be created, the request goes to the owner first. The owner checks the information, asks for anything missing, and gives approval before the record is saved. That approval step can be a paper form, an internal note passed by hand, or a short message by fax — the channel does not matter as much as the habit of not skipping the step. Whether the software itself can enforce this flow depends on which system the company is running.
The most immediate benefit of this model is the elimination of duplicate records. When the same supplier appears twice under slightly different names, the order history splits in two and payment tracking becomes unreliable. Once ownership rules are in place, checking for an existing record before creating a new one becomes a required step rather than an afterthought. The same logic applies to product codes. If one item is listed as ‘cotton shirt’, ‘men’s shirt’, and ‘shirt-cotton’ in three separate entries, stock reports will never add up correctly. A single coding standard, a single responsible person and a quick check before entry resolves most of this.
Ownership alone is not enough to keep data clean over time. Periodic audits are also necessary. Every three or six months, each owner reviews the records in their area. Inactive customer cards, outdated supplier addresses, product codes that are no longer in use — all of these get flagged and either corrected or closed. No special tool is needed for this. A simple list printed from the software is sufficient. The owner goes through the list, marks what needs to be changed, and updates the records. Once this habit is established, the master data pool gradually improves and reports become something the team can actually rely on.
Setting up this model in a working business starts with understanding the current situation. How many customer cards exist and how many are active? Are product codes consistent? Are there obvious duplicates in the supplier list? These questions can be answered with basic reports from the software. From there, an owner is assigned to each record type and that responsibility is written down. Entry rules — which fields are required, how names should be formatted, what information must be verified before saving — are written on a single sheet and placed where data entry staff can see it. There is no need to build a complicated system. Rules first, then habits, then regular review. When those three things are in place, the ERP software starts producing reports that the whole team can trust, and the constant back-and-forth between accounting and sales over whose numbers are correct finally stops.
This article was originally written in Turkish by Gökhan MERCANOĞLU on February 16, 2004 and has been automatically translated into English and other languages using machine translation.