The predictions for companies to win with data are not impressive. Gartner predicts that only 20% of analytic insights will deliver business outcomes through 2022, and that 80% of artificial intelligence projects in the same time frame will “remain alchemy, run by wizards whose talents will not scale in the organization.” McKinsey states that companies are “struggling to capture real value,” from analytics investments.
According to research by Harvard Business Review Analytic Services, only 18% of business leaders say they’re getting a sufficient return on their investment in analytics. The common roadblocks? Lack of standardized analytics processes and skills to interpret analytics results, siloed analytics and a time lag between when insights arrive and when they’re needed, the research showed.
Finally, 77% of company executives say that “business adoption” of big data and AI initiatives represents a “challenge” for their organizations, a recent survey found. “This issue, and the low percentage of companies that have achieved data-driven organizations and cultures, suggests the need for a new focus,” concludes the authors of the Big Data and AI Executive Survey 2019 from New Vantage Partners.
The potential of data analytics to drive untold economic efficiencies needs to be unleashed, and companies that do so will thrive. But as the NVP authors said, something new is needed.
History offers a valuable lesson here. Xerox PARC is credited with coming up with the first personal computer, but it took Steve Jobs and Apple to commercialize it. “Bell Labs scientists were responsible for the transistor, the solar battery cell, the fax machine, touch-tone dialing, the early communications satellite, improvements to radar and sonar, and much more,” Forbes reports. With more than 3,000 researchers in 12 labs located across six continents, IBM Research remains one of the world’s largest and most influential corporate research labs.
PARC, Bell Labs and IBM Research are all labs that fed product creation. Born of the industrial age, they created things like the PC and the fax machine. Today, data is flowing into and around companies, and everyone is trying to capture it to drive innovations not only in the creation and design of things, but of processes and services, too.
What more companies need today is a “data lab” to create ideas from data and a “data factory” to turn those ideas into products. Google, Amazon, and other data-driven giants already work like this. So should companies outside of technology.
6 Steps to Get from Anecdotal to Actionable with Data
Step 1: Get out of denial
This might mean you actually stop doing what you are doing even if it is a safe choice. Many companies today hire droves of data scientists and then expect products to result from the insights those data scientists create. What happens most often, however, is that everybody ends up frustrated and, as survey after survey shows, the return on investment doesn’t materialize. This is because data scientists alone don’t have skills to commercialize insights, and companies are failing in their current efforts to translate insights into products.
Step 2: Build two separate teams
The data lab team should be full of data scientists who question everything, believe nothing, and are ultimately tasked with coming up with crazy ideas that might just end up being the idea that makes your company super successful. Amazon was in the business of selling stuff when someone came up with the idea for Amazon Web Services, the cloud-services piece of Amazon that now drives much of its profitability.
A data lab could have thousands of scientists, like IBM, or two or twenty. Lab cultures need to be open and collaborative. Data scientists need great data tools, unfettered access to good data and backing from upper management in the face of frequent failures and dead ends.