The first part to this article titled “The Advantage of Good Business Intelligence” was published recently. This article continues from the conclusion that:
A Data Warehouse will provide the information that enables these questions to be answered via quantitative methods and analysis. Often this is provided by a combination of internal data and external research. The business will take the information contained within the Data Warehouse, analyse it, and use it to implement new goals, which in-turn require monitoring and further modification. This is a performance cycle that requires continuous monitoring for sustained success.
The ultimate use of Information is to create a competitive advantage for our business. The following picture recaps one from the earlier article.
The Business Drivers for BI Solutions
Leveraging Information for the Corporate Knowledge-base
Information is a competitive resource that is in increasing demand in the business environment today.
Information is made up from data, but data is not information. Information, when looked at in relation to a Data Warehouse is much more than a collection of data. Information has three sub-components, the past, the present, and the future. The task of turning data into information relies on the following primary steps:
◊ Extract relevant data from all relevant systems
◊ Transport data to the warehouse platform
◊ Stock source data tables
◊ Validate source data
◊ Identify timeliness of source data
◊ Assign consistent, repeatable keys
◊ Update dimensional attributes
◊ Update base measure attributes
◊ (Calculate derived measure attributes)
◊ Aggregate data using dimensional structure and business rules
Information is therefore processed data, associated with the structure of the organisation, its products etc.
One of the keys to generating information is tying data to a time-line. All data items will relate to a time period. Legacy systems cannot be relied upon to record when events happened; it is therefore the duty of the Data Warehouse to identify this.
Within the context of the Data Warehouse project it is necessary to recognise that information is a resource that must be managed effectively. Data cannot be an afterthought; it is the foundation for the project and must be effectively managed. This is the basis for the role of Information Resource Management (IRM) within the team. Data quality and integrity are also central to Information Resource Management. The objectives are:
◊ Planning and managing the organisation’s data.
◊ Documenting the data requirements and business rules necessary to support information needs.
◊ Defining an information architecture as part of the corporate IT architecture.
◊ Establishing an organisational structure to ensure effective data, quality and integrity management.
The following show some problem scenario’s concerning periodic (or time based) information. These are both based on real world scenarios encountered by the author.
Supermarket Example
Kostkutters 24 Hour Supermarket required a range of information about supermarket queues. Certain data they could ascertain from their own in-store data, e.g. frequency of use of checkout’s, the number of items each purchaser bought, Whether the transaction was paid by cash, cheque, or credit/charge card. Other data, for example the mood of the shoppers would only be available through the use of surveys.
Manufacturing Example
A component of turning data into information is the association of measurements to time. From these examples it is possible to see how time periods are an important factor in the decision making process. With the supermarket it is the time of day and day of the week that matters, whereas for manufacturing annual seasonality is important. Companies operating in the same industrial sector may have some common characteristics.
Historic Information
Historical information could be seen as the total of all history (potentially including yesterday). Normally historical information will be seen as data that is not part of any current time period. Historic data is normally heavily summarised, and is rarely kept at a granular level.
The amount of historic data to be kept will depend on the business being analysed. Some industries, like Banks, are governed by strict statutory regulations enforced either by the national governments or by international organisations. Before starting the Data Warehouse project it is essential to have an idea about how much history is
required.
Today’s Information
As soon as the project starts processing data from source systems it will become obvious that the warehouse rarely stores any information about today. If today’s information were actually stored within the Data Warehouse then it would be subject to change during the continuing business day. It should be noted that some current day events now have a legitimate part of some data warehouses.
In Data Warehouse terms today’s information is the current period, the current day, week, month, quarter, year etc. The term “Current Period” has special significance within many OLAP reporting tools and can be used in order to determine reporting periods.
The current period is the most recent period within the Data Warehouse. In designing the database it will be necessary to decide on levels of granularity and the longevity of the data stored.
Future Information
So we have built our Data Warehouse and filled it with current and historic data, so where do we get future information from? The Data Warehouse is the natural home of forecasting and budgeting applications.
Analytics & Business Intelligence
The population of the Data Warehouse is simply the enabler to provide analytical information, the Starting point for Business Intelligence (BI). Analysis is not simply the production of weekly or monthly reports. It is at a very minimum taking reporting outputs, identifying trends and measurements of key business importance.
Analysis is defined in the Oxford Dictionary of English as:
Detailed examination of the elements or structure of something, typically as the basis for discussion or interpretation.
and:
The systematic and critical assessment by an organisation of a feature of a product to ensure that its cost is no greater than necessary to carry out its functions.
Within the context of the business intelligence application analysis is the discovery of key performance indicators though the use of the information stored within the Data Warehouse.
Analytical use of a Data Warehouse relies upon the use of Business Intelligence or On-Line Analytical Processing (OLAP) tools. These are reporting tools with the ability navigate, through the dimensional structure and show the impact of important measures contained within it. Analysis of the information contained within the warehouse will contribute to the decision making process of the organisation. This is supported by the ability to drill down for more detail or drill up to higher summary levels.
Data Mining also aids analysis of the data. Data mining tools are a specialist type of analytical tool that provides the capability of applying mathematical or statistical processes for the purpose of extracting hidden knowledge within large data sets. Data mining techniques provide information inherent through patterns or relations within the data. This data can be used to make predictions about future behaviour based upon learning algorithms. Data Mining will be applicable once the organisation has built up sufficient validated history within its Enterprise-wide Data Warehouse.
BI Supports Better Decisions
Improving the decision making process is always the goal of any corporation. The term “Better Decisions” in the context of Diagram at the begining of this article means using the information contained within our Data Warehouse and applying relevant analysis in order to make informed decisions about the business. The decision is of course still a human action, what differs is that the person making the decision is making decisions based on facts, not on guess work.
The types of decisions made will differ according to the perspective of the decision maker. The following diagram looks at the differences between operational, tactical and strategic decision making. Each has a place within the corporate decision structure and must be supported by our Business Intelligence capability.
Operational, Tactical and Strategic Decision Making
Operational decisions are based on effective and efficient use of facilities and resources to carry out activities within budgetary constraints. Operational decisions are based on questions that include: What do we sell? Where is it sold? Who did we sell it to? The answers to these questions will generally be based on detailed (highly granular) data. Operational decisions are by their nature responsive to a changing environment. There will be a need for periodic reporting, such as the ‘Monthly Sales Report by Salesman’ but this is accompanied by ad-hoc reporting with new reports generated when they are required. Today’s demands will be different than next month’s queries. Operational reporting tends to have limited summarisation.
Tactical decisions are based on acquisition of resources, tactics the establishment and monitoring of budgets. They deal primarily with past events. Analytical tools are required to understand and act on the day-to-day changes in market demand, variations in environmental factors that may be unique to a specific industry or area of competition. Much of this type of analysis will be trend (or time series) based with a need to monitor how factors change over time, particularly with reference to industrial norms or in relation of specific competitors.
Strategic decisions are based on corporate goals, polices and determination of organizational objectives. Strategic decision making is about answering the following questions: What is the overall health of our business (now and in the future)? Are all the disparate parts of our organisation working towards the same goals? How many new customers will we have next quarter/year? How many will we lose to the competition? What new products/services will we introduce? What impact will technology, government regulations and customer patterns and preferences have on our industry?
A Data Warehouse will provide the information that enables these questions to be answered via quantitative methods and analysis. Often this is provided by a combination of internal data and external research. The business will take the information contained within the Data Warehouse, analyse it, and use it to implement new goals, which in-turn require monitoring and further modification. This is a performance cycle that requires continuous monitoring for sustained success.
Increased Efficiency
Increasing efficiency has been a long term corporate goal. All pay negotiations taking place today focus on increasing productivity, yet in the past productivity has been difficult to measure and has been dome very much on a hit-and-miss basis. The use of data within a Data Warehouse provides corporations with the opportunity to
measure organisational activities more effectively, especially when combined with applications such as Activity Based Costing (ABC).
A few areas where corporate efficiency may apply:
◊ Responding to budgetary constraints
◊ Increased import pressure as well know international corporations enter the local market
◊ Increased competition within the marketplace. This may stimulate productivity improvements
◊ Effects of privatisation on publicly owned corporations.
There are many theories about increasing corporate efficiency. It is not the place of this book to comment on those, it will be for the reader to put into practice whatever method they deem applicable. The writer is simply here identifying the role of the Data Warehouse in driving corporate efficiency on the basis of measured facts and not on
guesswork.
Uniqueness
Uniqueness is a rare business quality. The majority of businesses start life with the perception of having a unique product or service. But as the company grows their product or service becomes less distinctive. Sometimes the company name will also have a special place in the heart of many of its customers, which may extend the
distinctiveness of the product or service offered. However as we know business conditions change, particularly after patents and other intellectual property rights expire. Competitors enter the market with similar products or ones that are better than the original. Maintaining uniqueness is a difficult challenge, but a task that is made easier through the introduction of informed decision making.
Even where products/services retain their uniqueness improvements to efficiency are equally important in building a competitive advantage.
Competitive Advantage
Corporations are faced with the constant challenge of reinventing themselves due to either corporate takeovers or adapting to the competitive market place. In order to
react to these needs management need to:
◊ Have an in-depth knowledge of their company’s operations
◊ Know how key business factors affect the organisation
◊ Know how business factors change over time
◊ Understand and analyse their organisation’s performance relative to their key competitors or industry benchmarks.
In order to adapt corporations will have already been using Data Warehousing and Decision Support technology to measure what they are doing and will know the importance of corporate information. Gaining a competitive advantage requires using the knowledge gained from the Data Warehouse, identifying where processes can be made more efficient or unique in order to gain a competitive advantage.
Key business factors will differ from industry to industry from sector to sector from corporation to corporation but the need to measure and quantify results is constant in every organisation. Flexible solutions are necessary, to allow the company to re-invent itself on an ongoing basis in order to beat the competition. Corporations are therefore becoming more flexible with a cycle of change involving:
◊ Performance assessment based on information & analysis.
◊ Informed decision making.
◊ Training of managers, team leaders and team members to successfully implement new policies and working practices.
◊ Ownership of business events (including data quality)
◊ Identification and implementation of efficiencies within the business.
◊ Identification and enhancement of uniqueness.
◊ Continuous improvement to ensure the competitive advantage is maintained.
A Data Warehouse will provide quantitative methods and analysis to enable informed decision making. The business must take the information contained within the Data Warehouse, analyse it, and identify increased efficiencies within corporate processes of increase the uniqueness of the product/service offered. Alongside the corporate change is the need to ensure that staff at all levels are involved in the change process and understand the reasons for it.
