How to improve your Corporate IQ
7 steps to improving your Corporate IQ
Today, more and more companies are starting to realize that high quality information is a must if you want to make correct tactical and strategic business decisions. Numerous examples show that companies lose huge amounts of money, primarily due to poor information quality. No matter how much you invest in Business Intelligence tools - if you keep feeding them junk data, the output will be... well - less intelligent!
In this article we will describe how you can improve your Corporate Information Quality, by presenting a 7 step self-help program.
Step 1: Self awareness
Every self respecting self-help program starts with admitting you have a problem. You would be surprised if you knew the number of companies that are in denial about their low information quality. In other words, the first thing you must do is to stand up and say:
- Hello, my name is John and I have an information quality problem!
The first step to self awareness is to identify the pain in the organization. Find out which manager has nightmares because his critical business process is not working optimally. When you have found your manager you have found a potential sponsor for an information quality project. This step requires insight into the business. Although a consultant can help facilitating the work, somebody within the company is probably better suited to find the right candidate processes and managers.
Step 2: The value of information
A part of the problem when discussing information quality is that people seldom understand the true impact of low quality information. One reason may be that they do not realize the value of their information and therefore cannot understand the economical side effects of low quality information. On the other side, people that do realize the information value only tend to see it within their own silos. For a Call Center Manager, it is not particularly important if a name is spelled correctly as long as the phone number is correct, but for the legal department it is crucial to have the correct name of a client.
This step is about changing the mindset about how we view information throughout the organization. Compared to a company's key assets like money, people, buildings and equipment - information is, in many cases, not seen as a key resource and therefore not managed as one. You cannot change the mindset of the whole organization over night, and that is why it is important to have a sponsor. By proving the value of having high quality information for one business process, you have a platform for changing the mindset of other managers in the rest of the organization.
Step 3: Finding the Business Case
By now, you should have identified a business process that suffers from poor information quality. You should also have a manager and a sponsor, who is prepared to dig deeper into the problem. Finally, you should have a basic understanding of where in the process the low quality information hits the fan!
Now, let us take a look at the business case. With a business case you are able to see how to get a return on investment. At this stage, it is important to quantify the costs associated with poor information quality. Start by looking at the operational costs surrounding the business process itself:
- How many working hours are spent every year on reconciling bad data?
- How much do we lose every year because of lost order-invoice relationships?
- How many customers do we fail to invoice because of missing address information?
- How many unnecessary personal marketing campaign letters must we send because of duplicate customer information?
Continue by analyzing the potential business benefits associated with better data quality. Benefits could include shorter time-to-market, increasing customer satisfaction or expanding into new markets.
There is nothing more convincing than hard economical facts, when proving that you have information quality problems. If you do your homework, you will get management attention. In the end; would any manager say no to cost reductions and higher revenue?!
Step 4: Ownership
The fourth step is about ownership. Let's see...
- Who is ultimately responsible for the information in your company?
- The CIO.
- Is the CIO responsible for managing an organization that should provide clean and quality assured information to its business processes?
- Yes.
- Has the CIO appointed someone in charge of Information Quality Management?
- Hmm, don't think so!?
Ownership of information is sometimes debated. It is often said that the business process owners are in charge of the information that resides in their processes and systems, but too often it is the IT/IS organization that are appointed this task. However, the IT/IS organization is primarily interested in delivering a service to the business, in the form of integrated data loaded into one or several systems without any major hiccups. They are measured on the quality of the data delivery, not on the contents nor the data quality itself.
In most cases, they lack the knowledge of measuring and capturing more intrinsic information quality problems, like fuzzy duplicates or inconsistency and transposition errors.
Step 5: Establishing a Data Quality Function
A good way to start defining the ownership of information quality is to establish a Data Management Function. This is where a data quality process would be housed together with other data centric functions. DAMA International (The Data Management Association International) will release their Data Management framework during 2009 containing more information on how to setup such a function, check it out!
The Data Quality Management Function contains a vast array of activities designed to help the business improve their information quality. However, remember that trying to establish an end-to-end Data Quality Management Function from day one is the equivalent of a Big Bang approach.
Instead of trying to get everything in place at once, you should start by building the function just based on your recently found business case. In other words, start small but think big, and involve your sponsor!
Step 6: Use the right tools for the job
As you all know, the tools don't make the craftsman, but you would not build a house with just a screwdriver, would you?
Data quality software is great for creating awareness. With the software you can easily identify, quantify and categorize potential problems, like dirty data or broken business rules, and create statistics around them. Further along the data quality process you can use the software to cleanse and de-duplicate your data automatically before it reaches its target destination. In the meanwhile, you can monitor and control your data quality levels to see if they meet your targets.
To a Data Quality Management Function, the data quality software can be essential. Without it you may feel as a Security Manager without a firewall or updated antivirus software. Finding the root cause of one data quality problem is the equivalent of finding just one hacker!
Step 7: You can’t manage what you don’t measure
The last step to improve your Corporate IQ is to measure your progress as you go along. As the old saying goes "You can't manage what you don't measure". Measuring will give you the proof that identifying and solving information quality problems really has an impact on the bottom line!
At Affecto, we take information quality seriously. This is why we offer a wide range of knowledge and services to our customers within this field.
Dario Bezzina
The author currently acts as Solution Architect, Data Integration & Data Quality at Affecto Sweden.
[8.12.2008 08:49]
