Three common mistakes that can lead to data-mining failure
Enterprises are jumping on the big data bandwagon. They don't want to fall behind as rival businesses scramble to make sense of data they have collected -- for example, through new Web apps -- and turn their analysis into a competitive advantage. But watch out: "There are a lot of failure modes for this stuff" says Mason. Here are the three big ones:
--Over-investing in technical infrastructure. Many companies spend money on technology without knowing what they're going to do with it. "I have seen organizations spend millions of dollars on hardware and software without really a plan," Mason says.
--Hiring the wrong people. To turn their data into a competitive advantage, some business think all they need to do is hire a data scientist and end up hiring someone with math skills but no experience in actually translating business problems into math problems. After six months or so, they may end up firing the data scientist and wonder what to do next;
-- Failure in leadership. Businesses need someone who can cut through politics around who owns the data and what silo of the business is in control of it. Businesses need both strong practitioners -- people who will actually do the work -- but they also need somebody who is capable of giving them the power to get the work done. "I've never seen an effort succeed unless it had strong leadership," Mason says.
Sign up for Computerworld eNewsletters.