On Tuesday, International Business Machines (IBM ) launched three services to help companies sift and understand so-called "big data," hoping to get some of the $120 billion or more that businesses are expected to invest in data analytics by 2015.
IBM (Armonk, NY, USA), which has spent $14 billion itself on acquisitions to grow its analytics business since 2005, said it aimed to target top executive levels with its new offers of software and services to reduce fraud, manage financial performance and predict customer behavior.
It sees strong growth opportunities as companies, governments and organizations struggle to make sense of the large amount of data available. IBM offers the broadest portfolio in the data analytics market and competes with Teradata, SAP, EMC, NetApp and Oracle among others.
The idea behind big data is to converge structured data found inside databases and unstructured data found in social networks, mobile devices, meters or sensors to identify patterns or predict behavior.
The market for such services is growing at an annual rate of 60%, IBM said, and according to IDC research, companies will invest more than $120 billion by 2015 on analytics, hardware, software , and services.
IBM, which has a long history of analyzing data, began growing the segment almost a decade ago when the costs of computing started to drop. It now anticipates $16 billion in business analytics revenue in 2015.
"It's not just that we discovered more data could be good, it's become more economical," said Steve Mills, who heads IBM's software and systems unit. "There was really no computer science at that time ... all you had to do is show up in a suit, not say anything stupid and pass the DPAT."
Today, IBM's analytics can be found across various industries, from water and waste management to traffic and retail predictions. Its new services could be used to protect against insurance fraud, for example, IBM said.
Infinity Insurance, which specializes in covering high-risk drivers, was applying IBM analytics to detect patterns in historic data to identify fraudulent claims more quickly, IBM said.
In the healthcare sector, where Mills said fraud runs up to $650 million in costs per day, IBM systems use advanced algorithms to help spot fraud before claims are paid.
(Reporting By Nicola Leske; Editing by Gary Hill and Tim Dobbyn)