Data, like radioactive materials, has a half-life. That’s the conclusion drawn by a recent report from Nucleus Research, which tried to determine how long an average piece of data could remain valuable to an organization.
“The value of data diminishes based on the cadence of decisions,” read the beginning of the report. “Decision tempos are tactical (driving process changes in near real time), operational (driving changes that take days or weeks to implement), or strategic (driving changes that become part of a quarterly or longer planning and implementation process).”
Nucleus Research used data from 47 companies for the analysis, eventually concluding that the average “half life of data” for tactical decision makers is 30 minutes or less. “In actuality, the half-life of tactical decision makers had a high standard deviation, with some companies self-reporting a half life of as little as six seconds,” the report added. “On average, only 30 percent of remaining data has value after that point for predictive analytics and future planning.”
Things seemed a little more leisurely for operational decision makers, with an estimated data half-life of eight hours. Moreover, more of the information remained useful after that sell-by date: “On average, 30 to 70 percent of remaining data has value after that point for predictive analytics and future planning.”
For strategic decision-makers, the half-life of data stretched to an average of 56 hours, with a full 70 percent still viable after that point.
As the research firm points out, no organization relies on just one type of data; even companies with a strategic or operational focus necessarily need to devote a portion of the collective time to tactical concerns. Its analysis resulted in a chart of different companies’ decision-making “blend”:
In the end, Nucleus Research found that those organizations capable of increasing the tempo of their decision-making often had the most success in terms of setting themselves apart from the competition.
That seems pretty obvious—but the firm takes things one step further by suggesting that a strategy based around the concept of “data half life” can help managers refine data-capture and analysis strategies to a most optimal form. “Companies can evaluate the impact of the half life of data on their overall decision-making productivity,” the report concludes, “and optimize their data capture and analysis strategies to maximize returns on their investment in both technology and labor.”
Image: Nucleus Research