A New Data Visualization Paradigm?

Posted by Martin Karlowitsch on Oct 21, 2011 12:05:00 AM
A triviality to start with: significant data growth

It has been stressed over and over again, and there is no need to over-emphasize it here: every individual and every corporation both benefit and suffer from the massive data growth. They benefit because they can get access to more, more up-to-date and more precise information – about their customers, their markets, their suppliers and partners as well as their competitors and also employees and internal processes. However, they also suffer, as they have to deal with the increasing amount of data, have to manage and to store it, and have to enable decision makers to get fast access to only the relevant data. The more data available, the more analytics opportunities arise as people now can better slice and dice data, create segments, and drill down into every detail of their business. But along with this opportunity comes a risk as people often tend towards spending more time and energy on creating analytics than on making conclusions and taking decisions. This then causes paralysis of analysis. – A well accepted approach trying to overcome this paralysis is to culminate the analytics results into graphs. With graphs, tons of information and data points can get condensed into an intuitive overview and the theory is that this will enable people to make better decisions.


Data-driven decision-making and the need for speed

The importance of fueling decision-making processes with data visualization not only stems from the data growth phenomenon. I addition to this, decision cycles have become shorter and shorter. There are two dimensions to this: first, corporate decisions are both made for shorter time frames (many companies moved away from multiple or one year plans, and replaced this by and/or added quarterly and even monthly or weekly plans) and second, these decisions are made in faster intervals. The dramatic decision cycle decrease not any longer seems to be only valid for traditional high-velocity industries (like e.g. capital markets). The growing available data, the faster and more easy access to information, and the resulting overall increase in ‘market transparency’ has made many competitive advantages lose their relevance much faster; especially if these advantages were built on having an information edge. – Again, analytics data visualization seems to be a common approach to master this challenge as it enables decision makers to assess all relevant information at one glance. That way, decision makers do no longer need to spend time to collect data from various information sources and build their puzzle of the decision reality. State-of-the-art analytics dashboards deliver this in an easy-to-comprehend way, and as such allow making crucial decisions faster.


Decentralization of planning and decision-making

Until recently, many decision processes were centralized as this approach guaranteed speed and control. In these days, planning exercises implied looking at future projections and thus strategic, not operational data. This has changed simultaneously with decision cycles becoming shorter and shorter. The need for corporate agility created also the need to empower the owners of operational processes. These people no longer can be expected to just execute the given long-term strategy. They need to manage their “micro cosmos” self-sufficiently and need to be able to make process changing decisions on a day-to-day basis, e.g. rapidly reacting to market trends. This need has made arise all the “operational Business Intelligence” discussions during the past years. However, the point here is a different one. The gravitation center of the planning and decision process has been changing: it has become much more decentralized, and consequently it involves much more people than ever before. Well, as favorable as this is in terms of empowering people and increasing agility, it also has a downside: Not every “new” planner and decision maker is a data and/or a planning expert. – Hence, many tools had been created to enable everybody to understand the relevant data for their respective area of responsibility. You now know the point that I am going to make: visualization again has to play a significant role here.


So – what’s new?

Analytics and planning merge.

The answer to this question can be derived from a summary of the above discussion: Data volume grows, decision cycles shrink, and much more people are involved in success-critical corporate decisions. Ultimately, in such an environment the borders between analytics and planning processes get blurry: When operational process owners get their (graphical) analytical data points, they use it in a twofold way. They use the data e.g. to better understand potential deficiencies in their current process. At the same time, they also immediately turn this insight into corrective actions. What does that mean? – In operational analytics and decision-making, and this is where nowadays a key to competitive edge lies, there is no time left to analyze first, then decide and then adopt the planning. Analytics, decision-making and planning always used to go hand in hand. But given the developments that had been outlined above, de facto these functions started to merge.

So what?

Let’s reflect what we pointed out earlier: data growth, the need for speed and the decentralized decision processes all create increased emphasis on data visualization. Typical visualization techniques as applied in state-of-the-art dashboards and applications however are somewhat static. Of course, they allow powerful drill-downs and slice and dice operations, but they are static in a way that they just deliver insight to their users. Today, many data visualization approaches stop here. They do not allow the user to interact with the data: to modify it, to turn backwards looking analytics into forward-looking scenarios and to model out the impact of the data-driven decisions on the processes, and the to be expected results. Static graphics in the described way greatly support analyses and as such also decision processes. However, their usage potential for planning processes is limited. Planning never is static; it always must be interactive. Consequently, if analytics and planning merge, visualization techniques need to allow interactions between the user and the data. What many corporations started to need is real interactive data visualization with flexible Gantt charts.

This can be called a new data visualization paradigm.

What are your thoughts? How interactive do your planning and analytics data need to be?

Topics: Planning & Scheduling Insight, Gantt Chart Fundamentals