Analysis Paralysis: When Big Data Gets Too Big

July 23, 2018

Is your company suffering from analysis paralysis? Taabish Hasan recommends stepping back from the hype and applying a clear data management structure to achieve a strong, and sustainable, ROI. 


Size matters

According to Forbes Magazine, the world creates 2.5 quintillion bytes of data every day, and a full 90% of all the data currently rocketing around the planet has been generated in just the last two years. And this is only the beginning. As the Internet of Things (IoT) steadily connects disparate devices and tools, our collective data output will keep on booming.

Big data is often deemed to be an inherently powerful business force. Whether the data sets are drawn from internet searches, social media activity, or the operational systems underpinning most modern businesses—bigger is generally thought to be better.

Analysis paralysis

Reality, however, is more nuanced than the easy big data narrative. In fact, many executives eager to harness the strategic and tactical power of data analytics emerge in a state of analysis paralysis—smothered by reports and unable to decide what to measure, think seriously about, or act upon.

The Harvard Business Review (HBR) recently explored the corporate push towards data analytics, and the failure of big data to live up to the hype. HBR highlighted the fact that the percentage of marketing budgets assigned to data analytics is set to increase by a stunning 198% in the coming year, despite variable returns:

“These increases are expected despite the fact that top marketers report that the effect of analytics on company-wide performance remains modest, with an average performance score of 4.1 on a seven-point scale… this performance impact has shown little increase over the last five years, when it was rated 3.8 on the same scale.”

Lead the data or it will lead you

The challenge in analyzing massive data sets is significant. The volume of information at hand easily pulls the business in multiple directions at once, with meaningful, fact-based insights often lost within a multitude of general trend observations. Perhaps most importantly, drawing valid causal insights from many different correlations is a demanding task. If there is no central data approach in place, it’s all too easy for companies to misread correlation for causation.

For big data analysis to have an authentically positive impact, the business should be in control of the collection and analysis processes. Decision makers must, in other words, have a clear idea of what information they are looking for, before they start collecting and analyzing it.

Big data allows one to analyze pretty much anything, but the art of smart analytics lies in knowing what to analyze, and why you want analyze it in the first place. A common solution to the problem is to hire a top-notch data analyst, but this can be resource intensive, while also deferring a central challenge leadership will eventually need to tackle themselves.

Insights require data focus

A key factor in managing big data is understanding the variable worth of different measurement metrics at different times, and within different contexts.

For example, if a company has an average handle time (AHT) issue in one of its call centres, this will clearly be an important customer service metric to track. Once the issue is resolved, however, continuing to analyse AHT data wastes precious resources, and will likely muddy the proverbial waters. If the organization instinctively continues tracking such a metric, its capacity to perceive other important insights will be steadily reduced, while the intellectual manual labour required to manage the data analysis will grow ever more onerous as the reports pile up.

Thankfully, the solution to analysis paralysis is logical and process oriented. Instead of reporting across the full blizzard of possibilities that big data throws up, businesses should establish a basic analysis structure that focuses on a set of Key Performance Indicators (KPIs).

Conclusion

Strategically speaking, often the metrics we measure are secondary to the metrics we care about. And, because we don’t create the necessary associations between authentically meaningful metrics, we don’t see the full strategic story.

While large data sets offer the promise of important strategic and tactical insights, managing data flow and analysis can also drain company resources and blur strategic focus. Stepping back from the hype and applying a clear data management structure, rooted in business strategy, is key to achieving a strong, sustainable return on investment.


Taabish Hasan is a Director in the Performance Improvement practice at Farber. His practice focuses on developing strategies and action plans followed by successful implementations. Taabish can be reached at 647.796.6020 and thasan@farbergroup.com.