Data analytics grows but firms still put their trust in gut instinct

Data analytics grows but firms still put their trust in gut instinct

Summary: The increase in the use of data analytics tools is failing to snuff out the role of the traditional management hunch when it comes to business decisions.

SHARE:

Use of analytics software is rising but even firms that are the biggest fans of the technology still admit to putting gut instinct before data in management decisions.

Compared with 12 percent in 2009, 33 percent of organisations now use analytics as a predictive tool, according to research by consulting and services company Accenture.

Only 21 percent of organisations said they routinely use analytics very successfully as part of an integrated enterprise-wide approach

Yet despite increases in the use of the software, the firms most satisfied with business outcomes from analytics also said they are more likely to use intuition and personal experience ahead of data.

That pattern of behaviour is because analytics is not yet part of most organisations' DNA, according to Nick Millman, Accenture lead digital, data and analytics in UK and Ireland. "That's as much down to culture and the way people think rather than the capabilities of analytics currently," he said.

Enterprise-wide approach

Only 21 percent of organisations said they routinely use analytics very successfully as part of an integrated enterprise-wide approach. But the equivalent figure from 2009 was 14 percent.

Until firms stop confining analytics to one-off initiatives and embed it in the business, Millman said, they will fail to extract its maximum value and will continue to rely on gut feel and management instinct.

The research also flagged up a reduction in the internal resources available for analytical work. There are fewer analysts in individual departments — 68 percent of firms had them in 2009 but only 62 percent do now — and a fall in centralised analysts, down to 48 percent from 60 percent in 2009. More firms are now looking outside for those skills.

Millman suggested this change reflects the shift to predictive modelling from retrospective descriptive work.

"As organisations start to do more predictive analytics, you typically require a different skillset to doing descriptive analytics. [You need] a management scientist — someone who can build predictive models and think about the business insight that the model is going to generate," he said.

Chief data officer role

In the past 18 months, two-thirds of organisations have appointed a senior figure to lead data management, according to the survey.

"I've seen a growth in chief data officer role. I was faintly surprised that 66 percent of organisations claim they have a chief data officer or equivalent senior data leader at this stage. I think that's high," Millman said.

Only half of the organisations surveyed said they have consistent, accurate, formatted and complete data.

The researchers adopted a broad definition of analytics to include traditional business intelligence, newer predictive technology and data governance and data management.

"Analytics is the capability that turns data into business value by deriving insights from the data — those insights could be predictive or could be the more traditional insights from business intelligence, looking backwards at what has happened to understand the future better," he said.

How business analytics applies

He admitted that some types of decisions may not lend themselves to a data-based approach. "There will always be the need to make business decisions on the analytics but all business decisions are not equal in terms of how business analytics can be applied to them," Millman said.

"It might be a decision about how hard you should try to retain a particular customer when their package is coming up for renewal. Where you have a significant amount of data about the customer's demographics, their history with you, then I think you can make those very data-driven decisions," he said.

"You can micro-segment the customer base, work out who are the most valuable customers and then target specific offers at those customers you want to keep and you can do that with minimal human interaction in decision-making."

But those situations contrast with, for example, a strategic decision to acquire a company to break into a new geographic area. "You probably want to use data and fact-based analysis but there's always going to be a very significant human element to making the decision on which company you acquire," Millman said.

The research involved interviews with 600 executives in the US and UK.

Topics: Enterprise Software, Big Data, Data Management, Business Intelligence

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.

Talkback

2 comments
Log in or register to join the discussion
  • Gut feelings

    I guess some people just can't handle the truth...
    beau parisi
  • Incomplete Story

    Data Analytics, Business Intelligence, or whatever you're calling it is an incomplete tale. Organizations continue to see IT as their salvation for the "on-going concern." Yes, that's a Business term. Without Business, Business Intelligence cannot provide any useful insight. It's not enough to design a database, a data warehouse, or a dashboard without an innate knowledge of the Business behind the data.

    A query to determine the quantity, types, and time of day that customers buy your product is nice when you are able to equate that to the organization strategy. If the query results do not support the strategy, either your query is wrong or you're in the wrong business.
    Yangtze