Where is the data that we have lost in information? - TS Elliott
Where is the information we have lost in the business? - Simon Holloway
Here is some data: 38, 33, brown, 2. Do you know what I am talking about? Probably not. Let me now give you some context to this data, thus providing you with information - 38 waist, 33 leg, brown colour, 2 pockets. Do you have a better idea of what I am talking about? Perhaps, but perhaps not. Let me put you out of your misery if you haven't guessed yet, I am talking about a pair of trousers - men's more likely than ladies. But how do we want to use this information? To sell the trousers? To make them? To do an inventory check in a shop or warehouse? To find the trousers in your wardrobe? In each of these business processes this information is not enough. So when we talk about sharing data/information between applications (what DBMS was all about), what we have to understand is that each time we use the data/information we see slightly different views depending on the business process the data/information is being used in.
So to understand the data that we need to support our business we have to understand the different views that we have of certain collections such as Customer, Supplier, Employee, Product, etc. These views are based on the way in which the business processes wish to view or interact with the data. Additionally we have to understand who can see what data at what time in what location and on what type of device.
When I wrote the CCTA Data Management guide in 1994 , I introduced the concept of a lifecycle for data. For every piece of a "data collection" (a record, an entity) there is a lifecycle hidden in the business processes of your organisation. The phases of the Information Management Lifecycle are:
- Plan: To identify the requirements for information from users and business plans, and to plan for the acquisition, maintenance and use of that information;
- Acquire: To acquire the information identified efficiently and to a level of quality required for its use, and in a manner that meets the organisation's business objectives;
- Maintain: To maintain information so that it is kept in a state where it can be made available in the appropriate form, to the accuracy required, and in the appropriate time scale to support business objectives;
- Use: To provide information to the right people, at the right time, in the right place, and in the appropriate form to support business objectives;
- Disseminate: To inform business and IT users of the availability of information to meet their objectives; and
- Dispose: To remove information without compromising the integrity of the remaining information, such that there is an unacceptable risk to the continued provision of information to support business objectives.
Figure 1: Simple Information Life Cycle
You may have winced a little at this and said to yourself we don't plan for data. Well, actually, you do every time that you research and develop a new product or service to sell, or, if you are in IT, every time you have a new system, and personally when you buy a new car or house. The issue is that you do it subconsciously and this can lead to problems later on. So the tendency is for us to not really plan what data we need and how we use it. This results in unnecessary data duplication as well as a mountain of data being collected that we never use.
This covers both the process of creating the data initially and also purchasing data. The mantra is that we only want to create data once, but without proper planning and the amount of legacy systems that we have this can be difficult to obtain.
We also face issues that when we acquire data from external sources, we very rarely check its quality or that it has exactly the same meaning as the data with the same name, which will result in us comparing apples with bananas!
Certain data that we acquire is more sensitive and, in certain cases, we have regulatory requirements to report. Therefore we need to be able to identify which users or systems are able to create (and of course maintain or view) the data. This, of course, in its own right, involves a process to be able to associate rights of access to information and also to processes.
The rapid technology changes of this century have also meant that we can create data instances much more rapidly as we use automated data capturing devices such as RFID and Scart. This change in technology is leading to more systems being involved in processes as well as the need to sort through the information being generated so that only a certain amount is actually recorded.
But it is not only technology that requires data creation to happen more rapidly but also the business. In trading transactions, the speed is vital to the success of the transaction. In the automotive sector, you, the customer, are ale to change your mind about certain options up to certain moments in the production process.
Provision takes two forms: Use and Maintain. Once we have "acquired" data we then not only need to use it but also maintain it, as it is more than likely to change over time. The way in which we use data and information will be commonly through transactions that have been defined in the applications that we use to run our businesses. However, some organisations have made the move to a more event-driven approach, where a business event causes one or more business processes to be enacted. These are automated through the use of Business Process Management Software. Here the business process tasks are associated with either an application transaction that is presented to the user or a manual process that could be accomplished through an email or through a form which the user has to put the decision(s) on. In addition, certain decisions may have been implemented as business rules and stored in a Business Rules Engine.
It is no good acquiring data, either internally or externally, if you don't let people in your organisation know what data there is. This means data must have definitions, sizes, formats, etc so that people understand what it means, how it can be used and, most importantly, where it is being used. This gets into a big discussion area of data ownership or what I prefer to call data stewardship, as no one in an organisations "owns the data" - it is the organisation's. In the past we used data dictionary systems to record this, they then became repositories and now they are master data management databases.
This is a part of the lifecycle that we all too often forget about. Therefore we end up with loads of out-of-date data that we should have either archived or deleted. All data has an end life. Sometimes this is controlled by government or industry regulations, sometimes by organisation policy: whichever it is there is a process that needs to be employed that allows the data to be removed from active, and even archive data, storage and disposed of. In certain circumstances this process and the tasks involved will need to be recorded to conform with regulations.
Data and information are the lifeblood of any organisation but without properly defined business processes that allow the raw material to be created, extracted, analysed, manipulated, shared, maintained and disposed of, it is useless. Business processes are triggered by events, normally from external sources, but not always. Some of you may have heard the term "Enterprise 3.0". The difference between 2.0 and 3.0 can be summed up as follows:
- Information has moved from being static to dynamic in nature
- Processing has moved from transaction-based to event-based
- Storage of data has moved from database to ESB
- Applications have moved from ERP to BPMS-based sitting on top of legacy applications
- Business intelligence has moved to real time business rules
- From a 2 dimensional world to a 3 dimensional one.
All this is so that we can support he business need for agility and responsiveness to changes.
 Data Management, CCTA, Information Management Library, 1994