Wikipedia devotes just under a page to the subject of Master Data. A surprisingly small amount given the crucial role Master Data plays in managing profitability in business. It doesn't matter what software you are using - get your master data wrong and you can watch your profits walk out the door and wash down the drains.
I know this, I've seen it happen.
First, let's talk about what master data is - then we'll talk about why it needs to be accurate and well managed.
According to Scott Taylor (a gent who styles himself as a Data Whisperer) "Master Data is a single source of common business data used across multiple systems, applications and processes." Great - that tells me what I already knew; it gets used everywhere in the business - but what exactly is it?
SAP divides Data into two basic categories:
Master Data, and;
To understand how SAP manages data it's best to think of SAP (especially in a whole of business application like ERP) as a virtual version of the business it is working in. It works in a similar way to a tactical operations map which creates a virtual battlefield - except it uses numbers instead of symbols. It parallels physical assets, people, processes, purchasing, production, distribution and sales. For example - every real person working for the business exists as an entity in SAP, every building and fleet item exists as an entity in SAP. Every item purchased and placed on a warehouse shelf exists in SAP. Every time a customer buys something, SAP parallels the process - right down to picking the stock off the shelf and managing the payment. Labour time spent to do all of these things is tracked in SAP. It's not a precise analogy, but it is a good place to start.
Master Data more or less mimics the physical and actual (or fixed) things - assets, fleet, stock, people, suppliers, and customers. Most importantly, master data collects and tracks the dollars $$$$$ - General Ledgers and Cost/Profit Centre Datasets are always Master Data - they do not change.
Master Data includes (but is not limited to) the following datasets:
Transactional Data roughly parallels processes (makes something happen with, or to, Master Data) - eg the process of buying an item, selling an item, making an item, moving an item, breaking an item, hiring someone, transferring someone, firing someone or paying someone. Transactional data moves the dollars $$$$$ around.
Good management of master data is crucial to the profitability of a business. It all boils down to ensuring that the data matches reality as closely as possible - which is not nearly as easy as it sounds. The real world is involved, and in the real world the numbers do not always line up perfectly.
Let's start with stock values and purchase prices. In the military ammunition is always a stocked item. It is assigned a stock (or barcode) number which does not change regardless of where the ammunition goes in the organization. Unless the Master Data is maintained in accordance with real world price rise cycles the stock value given to the item in SAP Master Data is almost never the same as the price paid for it on purchase. We’re going to leave fixed price purchase contracts aside for the sake of a simple explanation. Lets say a 7.62 round is assigned barcode number 5551555. The master Data in SAP for the 7.62 round might look something like this:
5551555 AMMUNITION SMALL ARMS 7.62 NATO LIVE $0.20 Qty 1
When the purchaser pays for the ammunition they might accept a price rise of 5c per round - giving the round a real world value of 25c. When the round is accepted into stock and valued SAP will still value the round at the barcode value of 20c - the Army pays 25c for it. Immediately there is a difference of 5c per round between theoretical or planned stock value and the actual commercial value of the round. If the Army buys a million rounds there is an immediate dollar value variance of $50 000 between the planned value of the stock and the price paid for it.
The process of revaluing the Master Data is not always automatic. SAP will not automatically see the difference between master data value and purchase price and automatically update (unless you have set it up to do so). A person has to initiate the process of revaluing the master data. Revaluation does not always happen in real time. Unless the procedures are watertight and strictly adhered to, master data valuations can often be over five years old.
When SAP reports on stock value & usage for future procurement planning it draws the Report from the Master Data valuation - not the Purchasing Spend Transactional Data. If the master data is not up to date and the person doing the planning does not look at the purchasing spend, the planned spend on restocking might as well be a fairy tale.