Frank felt he ended up in episode of ‘Little Britain’, “the computer says; no”. He reruns the same report again, expecting a different outcome by slamming his keyboard harder, but the computer is adamant. Defeated by a machine Frank hauls himself over to the IT department, scraping together the courage to do another ‘5 Whys’ routine with someone as inflexible as the computer, but hopefully a little more empathic.
To Frank the information system is a black box that magically produces numbers on which he makes his management decisions. There is just no substitute for the system, harvesting information from planning, production, resource, sales, purchase, forecast; the automated filtering, weeding, welding of all the pieces; it just couldn’t be done by hand. At least not on a day-to-day basis. But from time-to-time Frank just knows the numbers don’t add up, sometimes literally. In such an event the IT guys have to dig, compute and analyse; and within a few hours they will be able to tell Frank what exceptional situation causes the phenomena he discovered. And after a day or two the blackbox will have an additional rule to prevent the situation of occurring again. Sometimes Frank wonders; how many times did he make decisions without ever noticing the information wasn’t correct; could the blackbox ever get so complex that nobody could understand what is was doing any more?
To prevent Frank of waking up in the middle of the night, worrying too much, his IT department must analyse -the- five key components on which any information system is build. By analyzing these five components for the various parts of the system they will be able to drastically improve the system reliability. Although we will look at the five components separately, they work together in synergy. The quality of the information is not the sum but the product of these five components. So if you want to improve on your information quality you have to enhance at least one of these five without letting the others drop.
Factual. This looks trivial, but on a day to day basis you take into account lots of information that simply isn’t correct. Most of the time this is caused by the route the information has travelled to get to you. Every node in this route will alter the information to some extend. The more nodes the more scrambled the information gets, like in the ‘Chinese whispers’ game. Shortening the route is not always an option, creating additional alternative routes almost always is.
Actual. Once information reaches you it is at best dated but very likely outdated. In most cases information is conceived from older bits of information which, among each other, may vary in age as well. You increase actuality by understanding how information is created. In many cases it is hard to tell factual from actual, don’t worry, you have to improve them both.
Applicable. To make a sound decision you need good information and more information is usually better, but more information will costs more time to analyse. To save time we have computers to aggregate and condense large amounts of data into simple KPI’s. But rare occasions do occur and wrong input can have devastating effects. A good information system won’t be able to handle these exceptional circumstances but it sure can detect them. By setting sensors on your input, you are able to detect out-of-bounds information.
Integral. If you receive information you have to be sure that you aren’t missing out on essential bits to complete the picture. An information system however cannot analyse every alternative, it has to exclude large amounts of information at forehand. The trick is to teach your information system when output is out-of-bounds. Applicable and integral might be the same when looking at an intermediate step of an information system, however it is wise to use a different approach in detecting out-of-bounds output, than for detecting out-of-bounds input.
Traceable. If you are relying on information that is given to you, you must have an insight on how this information has gotten to you. A proper trace from information source to information target is essential in a healthy information system. You must regard this as you health policy, if you don’t take care of it now, your information system will get sick.