December 15, 2015

Another Survey Questions Value of Big Data

George Leopold

Enterprise data crunchers still have little to show for their efforts despite huge investments in collecting, storing and analyzing large datasets, according to a new “data chaser” study that also found that nearly one-third of users fail to act on collected data.

The survey released this week by Square Root, a software-as-a-service and management software vendor based in Austin, Texas, found that nearly one in three companies fail to take meaningful action based on their analytics initiatives. Moreover, nearly half of those companies surveyed said their employees’ time would be better spent elsewhere.

Part of the problem is a lack of integration along with outdated tools. Manual tools like spreadsheets are still widely used, and 70 percent of companies said they are relying are at least three different tools to crunch data.

The upshot, according to Square Root CEO Chris Taylor, is that “simply chasing down large volumes of information can lead to decision paralysis and waste both time and money.”

Worse, the survey found that more than 55 percent of respondents admitted manipulating data to reach a desired outcome. In the retail sector, for example, half of those executives surveyed acknowledged cherry-picking data. Overall, half of managers surveyed said they made decisions that ran counter to the results of data analysis.

The financial services sector, long viewed as a driver of IT and analytics technologies, paradoxically was the sector most likely to collect data and then ignore it. One reason, the survey found, was that a growing number of data sources made analysis too time consuming.

Based on the survey findings, Square Root asserted that the key to moving from “data chasing” to meaningful results would require a focus on delivering timely information and “just-in-time” insights.

While companies like Square Root continue to pitch data management and other software services, database vendors have embraced technologies like graphics processors to help accelerate data queries that previously held up the process of gleaning insights from soaring data volumes. The problem has grown more acute as the amount of unstructured data expands exponentially.

In-memory approaches used to speed up database queries are slowly giving way to GPU-accelerated graph databases and machine learning tools that can be used to automate the most time-consuming aspects of data analysis.

For example, predictive analytics and machine learning are increasingly being used to parse new data streams, but query times are slowing as data volumes soar. Hence, there is a growing emphasis on technologies like GPU accelerators as a way to improve performance and speed database response times.

Besides technology, the lack of data traction in many organizations may require a cultural shift. As observers have noted, the hype surrounding big data has prompted many executives to conclude that they need a big data strategy even though they are unsure about how to use the capability.

Or as an Intel Corp. executive noted earlier this year: “The dirty little secret about big data is no one actually knows what to do with it.”

Indeed, the biggest barrier to adopting big data technology, according to an earlier survey by market researcher Gartner, is determining its value to an organization. 

Recent items:

Intel Exec: Extracting Value From Big Data Remains Elusive

Harnessing GPUs Delivers a Big Speedup For Graph Analytics


Share This