The data industry's next task: turning customers' instincts into useful information

SAN FRANCISCO — The data evangelists’ mission is half complete.

Every modern company wants to use its data to cut costs, increase revenues, and save time. But the hard part remains: helping companies do exactly that.

[aditude-amp id="flyingcarpet" targeting='{"env":"staging","page_type":"article","post_id":1476392,"post_type":"story","post_chan":"none","tags":null,"ai":false,"category":"none","all_categories":"big-data,business,","session":"B"}']

In a panel at VentureBeat’s DataBeat conference this afternoon, executives from four data startups — Ayasdi, Clearstory, Databricks, and Interana — agreed that their customers have the right gut instincts on how to use data to benefit their companies. But the right insight doesn’t always equate to valuable findings.

Unsurprisingly, they had lots of different ideas on how to best bridge that gap.

To find relevant data, Clearstory CEO Sharmila Mulligan thinks it’s essential to tie together data from multiple sources — separate business divisions and actual storage locations — into a single environment. This tends to show important contrasts and comparisons that may otherwise remain hidden, she argued.

As those data sources grow, Ayasdi cofounder Gunnar Carlsson stressed the importance of a flexible environment for data. If there’s a representation that accepts all data types and produces something you can easily interact with, solutions often materialize in front of your eyes — even if you’re not aware of an initial problem.

To illuminate his example, Carlsson highlighted Ayasdi’s work with a large hospital system. Ayasdi analyzed the hospitals treatment patterns — timed lists of various interventions doctors and nurses perform for patients — and discovered a lot of inconsistencies in the way they treated patients with the same conditions. Ayasdi built a clear representation for its customer, which began to reconcile those differences across their various locations — and saw major cost reductions.

But there’s some onus on the companies to make sure they’re well staffed with domain experts, said Databricks CEO Ion Stoica. These folks, ideally people familiar with data practices, will be able to not only understand a company’s issues but also what data is relevant to finding a solution.

Yet that doesn’t require a complex approach, stressed Joe Adler, Interana’s director of product management. It’s rare that companies need to open up the toolkit and advanced statistical methods to realize significant gains from their data, he said.

“We have a tendency to complicate these things, but asking simple questions about data, going through that initial summary, [and] watching data over time gives you so much insight, so much ability to run a business better,” Adler said.

[aditude-amp id="medium1" targeting='{"env":"staging","page_type":"article","post_id":1476392,"post_type":"story","post_chan":"none","tags":null,"ai":false,"category":"none","all_categories":"big-data,business,","session":"B"}']