To paraphrase Peter Drucker, information technology tends to focus on the technology, not information.
Instead of focusing on more data collection, greater storage capacity or faster data transmission, the objective of IT needs to be giving people the information they need when they need it to make good decisions. Information presentation exists, but “dashboards” and reports may not yield the information people need or give it in the format they can readily make use of. This is where delivering data in an easily digested format is often mistaken as the key deliverable, ignoring the fact that the report may not be what key decision makers actually need.
Sometimes this mistake is compounded by collecting more types of information in the hope that it can be made useful. Big Data experts are well paid precisely because we gather too much information and then try to figure out what we need and how to deliver it so that knowledge workers can use it to make decisions they need.
According to Peter Drucker, lots of executives are so busy counting this and analyzing that that they forget that any measurement is meaningless at best and counter-productive at worst if it is not done solely with the goal of helping the organization meets its mission. Collecting information you don’t need, analyzing it when there is no need to do so and pouring over reports looking for the answer to a question instead of having it provided on demand are all types of wasted effort.
What do users need? Think of the statistical process control charts that give a machine operator a clear warning that something is wrong and that they need to call for service, versus a screen full of metrics and trying to figure out if much less how to act.
When there are reports with useful information that isn’t readily available, alter the reports or standardize their transformation so that there is a simple and consistent delivery of answers. The only thing worse than someone having to analyze a report to get the answer is wasting time running the analysis and acting on the wrong information because of a mistake they shouldn’t have had the opportunity to make in the first place.
Industrial engineers need to be involved in the process of delivering information to users. You can’t rely on technical wizards to determine what data to collect and report. I know from personal experience that their goals focus on their own area of expertise. Statistics like website views, system uptime, network speeds and data loads matter more than whether or not one key knowledge worker received the right report on time or the resources invested in converting exported reports into the final form other knowledge workers need to know what to do.
There are also times where the metrics IT wants to deliver aren’t the granular pieces of information that are needed to improve operations. For example, the 97% customer satisfaction rate won’t give you the specifics on the 3% who were unhappy, much less their demographics or the details of their cases necessary to make them happy. Statistics on how many level 1 outage tickets versus number of level 4 enhancement requests came through don’t give managers the suggested improvements to reports or user interfaces that would have a strong ROI if implemented as the next IT process improvement project.
The solution isn’t yet more data to wade through but smarter, more efficient data collection and analysis.