What are the benefits of a data warehouse?

Most IT systems are designed to record the day to day transactions of a particular part of an enterprise. Your accounting systems will record cash flows but not stock levels or customer interactions. Enterprise Resource Planning (ERP) systems attempt to cover the recording and processing of information enterprise wide and consolidate all corporate information into a single database. ERP systems are primarily designed to record and process transactions and have limited and often simplistic reporting ability. 


Another problem with ERP systems is that they are completely unable to cope with the different reporting paradigms that are required for actuarial and marketing reporting. A quick glance at most operational systems will show that reporting needs are mostly an after thought and when available are either extremely rigid or hopelessly complex. When reporting is too rigid you may only be able to retrieve a fixed report which partly covers your information needs or you may have to spend excessive time and money in having a purpose built report created for each employee. Some systems give you a set of tools that you can use to interrogate your operational system but to use these usually requires programming skills and a good knowledge of the IT side of the operational system. 


Most business managers simply need the information, they do not have the time to become programmers as well as managers. Even if they did, they would probably have to learn a new tool for each of the operational systems in use within the enterprise. In order to minimize the amount of time spent searching for knowledge and maximize the amount of time actioning knowledge, an enterprise needs a system designed with reporting and summarizing knowledge enterprise wide as its primary purpose. A data (or information) warehouse is one such system. There are many benefits to owning a data warehouse and we have listed some below.


Productivity reporting is possible. Reporting systems attached to operational systems usually have a limited ability to perform summarization and comparison operations. While reporting from an operational system may allow you to view total sales by region, it is unlikely to allow you to view profit by region and by product for any specified time period.


The operation of calculating profit (selling price - [cost + expenses]) can be extremely complex especially when infrastructure costs must also be accounted for. A well designed and implemented data warehouse can automate this calculation and allow management to review profitability at all levels on a daily basis.


  • Queries made against the warehouse do not impact on the operational systems. A complex query of your operational system can be very resource intensive, and this may slow the operation system and will impact on response times of mission critical users. In a real time system a complex sales report run by your marketing manager may slow your systems to the point where telesales have difficulty punching orders. This leads to both frustrated employees and angry customers. If reporting is done from a warehouse there would be no impact on the transactional system or its users. 
  • The same set of data can be viewed in different ways and at different levels by different groups of users. Marketing may be interested in viewing the number of sales by sales rep. and Accounting may only wish to see the cash value of sales by cost center. If both groups obtain their information from different systems, there may be no correlation between their results which could lead to conflicting viewpoints and actions. By using the same source data we can be sure that different views of the data are legitimate. The level at which information is presented is also relevant. Your financial manager may wish to see all transactions and be able to drill down to the source data while a regional sales manager may only need to see sales information for his region. Again the different users must be using the same underlying data if their answers are to reflect a true picture of the enterprise.
  • The accuracy of your operational system can be checked. Errors creep into the data stored into your operational system only too easily especially, if data validation at the point of data entry is not absolutely rigorous. The process of building a warehouse must include checks which detect bad data, and this information can then be passed along to the people who maintain your operational system for correction. If an operator fails to enter a customer number and the operational system does not detect this, the transaction may be lost within the operational system and the downstream effects can be complex and costly.
  • Queries that require information spanning several operational systems are possible. As an enterprise grows it adds new systems to supply it's information needs. Often these systems are incompatible, or information stored in one system is only able to be related to information in another after complex and time consuming manual manipulation. It is easy to arrive at a situation as shown in the diagram below. The diagram shows how a mixture of reporting requirements, all independent of each other, can arise within an enterprise. In a data warehouse, data can be imported in virtually any format and the manipulation can be automated so that data in previously incompatible systems can be compared and reported on. • Staff spend less time searching for knowledge and more time acting on the knowledge they have. Some estimates indicate that senior managers can spend up to 80% of their time locating and manipulating information into a usable form. These people are paid to make decisions and act on these decisions. By providing the knowledge an readily accessible form from a data warehouse, you give them more time to do what they do best - make decisions and act on them. 
  • The exchange of business information between different enterprises is possible. In different operational systems data fields may have same name but record different information thus making comparisons or exchanges of data impossible between different enterprises. For example, in one enterprise a general ledger may record business date as the date the client paid. In another system business date may reflect the date on which the client signed the contract and payment date may reflect when the client paid. The warehouse building process will establish the correct relationships and business rules to ensure that data is comparable. 
  • The information can be obtained using any number of tools that the end user is familiar with. When asking complex questions of data within the enterprise, the choice of questioning tool is as important as the availability of information. It is well known that people will use use lower quality knowledge resources simply because they are more convenient to access than a higher quality knowledge resource. Data warehouses allow a wide variety of interrogation tools and it is possible to select a tool that your staff are, at least in part, familiar with rather than being forced into using some arcane proprietary tool supplied with your operational system. This means that your staff will find it easier to ask the questions they need to ask and to get better quality answers.
  • Customers can be provided relevant information via the enterprise eCommerce site. Operational systems are seldom designed to deliver information to the internet and the ability to exchange information between your eCommerce web server and your mainframe based operational system may at best be poor. Data warehouses act as valuable link between the two systems and can act as the foundation for your Customer Relationship Management (CRM) efforts. A data warehouse that stores client transactions may serve to feed a web site that allows your clients to view online copies of their orders and the status of those orders.
  • Interactive Voice Response (IVR) systems can be provided with data. An IVR system can function much as eCommerce site does but can reach those of your customers who do not have access to the internet. Again the data warehouse acts as the link between IVR system and operational system.
  • It becomes possible to have a single view of all interactions that a customer has with the enterprise. Too often different operational systems do not speak to each other and when a customer has a query they are shunted from one department to another. A warehouse makes possible the ideal of a "one stop shop" call center since the warehouse has brought information from all the enterprise systems into one place. This means that any information about the customer can be retrieved and used to rapidly resolve customer queries. The value of this improved service is immense.
  • Customer segmentation and niche marketing become possible. Because of the value added by metadata to the data warehouse and the summarization that the warehouse process employs, it is possible to quantify and identify discrete segments of your customer base and ask questions about their purchasing habits. Without a data warehouse, asking these multidimensional questions is either impossible or requires expensive purpose built software.
  • The end user of the knowledge stored in the warehouse is empowered to retrieve and manipulate information relevant to them and their role within the enterprise. With reporting tools such as the Naxian Viewer or the Naxian web reports, it is the end user who decides what information they need which they can then retrieve. This means that the user gets exactly what they need, when they need it. The responsibility to develop a report portfolio for each user is taken off the IT department and given back to the end user who decides what information they need and how they want it filtered and displayed.
  • Reporting periods and business cycles are not dictated by the technology. Some ERP systems are designed around a monthly business cycle and when deployed in business that require weekly reporting , they are unable to deliver the necessary reports. In such circumstances a warehouse can be used to provide reporting that matches the business need while the ERP system can be used to do what it does best, transaction processing.These are just some of the benefits of a warehouse. From this we can see that data warehouses serve the reporting needs of an enterprise far better than any reporting system or tool attached to a system whose principal purpose is to carry out transactions.  


"If you can do X better than you can do Z, and there is a second person who can do Z better than he can do X, but can do both X and Z better than you can, then an economy should not encourage that second person to do both things.

You and he (and society as a whole) will profit more if you each do what you do best" David Ricardo (1772-1823) Ricardo's Law of Comparative Advantage

To follow...
  • MTN
  • Santam
  • Absa
  • Old Mutual
All Our Clients
Case Studies
  • Old Mutual
  • Santam
  • Absa
  • CSC
  • Uskotec
  • MTN
  • South African Airways
  • Vyinde
More Casse Studies
The 500 Year Delta

"The Information Age...is dead and gone, replaced by the information economy...The Information Age was about the building of databases. It was about the rise of…

Read More