IBM launches business analytics services unit; Eyes predictive modeling

IBM launches business analytics services unit; Eyes predictive modeling

Summary: IBM said it is pushing for "fact based enterprises" that operate more on analytics than intuition. In other words, gut feel will be downgraded in management. Big Blue also noted that it expects rivals to follow.

SHARE:
TOPICS: IBM, Software
5

IBM on Tuesday unveiled a new services unit focused on business analytics and optimization and stumped for the need to use data to make better predictions.

At an event in its Hawthorne, NY research facility, Big Blue detailed  plans to launch its first new services business since 2002. In a nutshell, IBM is taking its knowledge of verticals, mathematics, research and optimization software and lumping it into a service designed to "improve the speed and quality of business decisions."

The move (statement) also wraps a service around recent acquisitions such as Cognos and iLog and links the technology to IBM's PwC Consulting unit. IBM is pitching its latest shift as a "foundational change" that will intertwine analytics in business management. The goal from Big Blue's perspective: See around business corners and minimize risk.

Ambuj Goyal, general manager of IBM's information management software group, said the company has been building out its assets---software, best practices and research---to offer this analytics service.

"How do we use information as a strategic asset? Human capital is a strategic asset. Cash is a strategic asset. Companies don't think of information as a strategic asset," said Goyal.

IBM said it is pushing for "fact based enterprises" that operate more on analytics than intuition. In other words, gut feel will be downgraded in management. Big Blue also noted that it expects rivals to follow.

Here's the quick summary of what IBM sees:

Executives such as Frank Kern, senior vice president of IBM's global business services unit, and John Kelly, director of IBM Research, talked about the rollout.

Kern said that businesses need to become more predictive and deploy "enterprise intelligence." "If we would have had systemic risk predictive capability and looking at risk and how it compounded, how would have the world have looked different today," said Kern, referring to the financial industry. "You can take these questions for financial services, security, health care, water. How can I get new insights on threats based on the data available to me?"

"Decisions are being made on gut feel and intuition. That can't be the case going forward," said Kern.

So far, there are 4,000 consultants dedicated to the new services unit along with 200 math wonks and researchers. Kern said that the services unit "will grow from there."

The scheduled demos highlight a few areas where this service will be used. IBM is set to show off a fraud and abuse management system, inventory management tools and a crime information warehouse.

"It's a rarity that you create a new line in the consulting business. This is not something we take lightly," said Fred Balboni, head of the business analytics and optimizations services unit.

Balboni noted that businesses need to move away from "read and react" to be more predictive.

IBM's latest service is built on:

  • Strategy, governance and process improvement;
  • Business intelligence;
  • Analytics and optimization;
  • Enterprise information management;
  • And enterprise content management.

Add it up and IBM has the key parts. But IBM is essentially selling ways to cut costs, increase profits and improve efficiencies. The big question is whether IBM's analytics service will be paid based on those outcomes. Balboni's examples included knowing when a top 2 percent customer calls a call center; adjusting peak loads for a utility; and dynamic inventory replenishment.
Goyal said that IBM will focus on a bevy of short-term and long-term ROI metrics.
"The fact-based enterprise is a different beast. They change the way they think. We're talking about a fundamentally different company," said Balboni.

Topics: IBM, Software

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

Talkback

5 comments
Log in or register to join the discussion
  • All watched over by machines of loving grace

    IBM seems to want to remove the human element, which is the only infinitely variable part of a business. By opting for safety and predictability, you create institutionalised mediocrity.
    Patrick Neylan
  • RE: IBM launches business analytics services unit; Eyes predictive modeling

    Before anyone gets too excited about Algorithms being used to automate business. Where do you think IBM tested this formula??? Well on themselfs of coures. Results? At the encouragement of IBM more than 40% of workers work out of their own homes. Well IBM just anounced that, as of May 1, 2009 they will NO LONGER PAY FOR INTERNET ACCESS FOR HOME BASED WORKERS. This is the lifeblood of comunication between workers! IBM answer... Go to an IBM office to work.
    Sounds simple, right? Well much of the Brick and Morter offices are long gone!
    What else? Two layoffs in the USA and one coming in Europe. This is a profit making company!

    Algorithms at work!!!!

    Bobbear
    Bobbear
  • RE: IBM launches business analytics services unit; Eyes predictive modeling

    I am a young Practicing software engineer (systems) at a Mobile Applications Provider Company in Kenya, freshly graduated in Computer Science at one of the local Universities.

    I am so fascinated about IBM's new product on one had as a passionate computer scientist and on the other hand as a computer scientist with passionate interests in computational modeling, operations research, optimization problems and algorithms. I highly second IBM's innovation and the move to let businesses change their current methods of work to what I can call "Intelligent Business" which is what IBM is supporting.

    Congratulations to IBM.
    mulongo
  • Another panacea...

    We've been hearing for decades of how the next big thing in engineering/ computing would create an "entirely different kind of company."
    Take a peek at the posts from Larry D's other article... The posts on MRP make for interesting reading:
    http://talkback.zdnet.com/5208-10532-0.html?forumID=1&threadID=63305&messageID=1174632

    Check out G?del's theorem - there is NO silver bullet! No algorithm or set of algorithms can solve all problems. It's officially and mathematically impossible. Therefore, IBM's and Microsoft's dissipated "solve-all-problems" solutions may earn money from bullish and ignorant CEOs who do deals on the golf-course... But the smart money will go to human consultants who actually understand which algorithms to apply in which situations, and can recognize a situation that requires the development of a new approach.

    I was lectured on computer security at Cambridge University, by Professor Ross Anderson; who is highly respected in his field. Fraud is innovative. Fraudsters are creative. They look for the gaps in the system. There are gaps in every system.
    matthew_of_cambridge
  • Another panacea #2...

    Let's look for one minute at IBM's claims (as expressed in the article...) Problems:
    1. They're inventing standardized methods of analysing risk, including the risk of fraud... (Whatever methods people invent, someone will always find a way around them. If you standardize the fraud detection/ prevention, you can standardize the fraud and criminals can teach each other in prison how to do it.)
    2. IBM's analytical techniques are based on analysing existing companies (this is the scope-of-experience of IBM's best developers)... By applying their techniques, their system will create "an entirely different kind of company." Anyone see a problem with this?

    If anything, IBM's methods will introduce NEW methods of committing fraud... Imagine the crimes that could be committed by feeding spurious data into an IBM decision-making system (or indirectly by manipulating the particular "facts" on which the decisions are based...) I could analyse the techniques IBM has applied, and then find some means of manipulating them. Imagine the chaos from large numbers of enterprises simultaneously imploding because of having all made the same mistake (due to all of them employing the SAME somewhat flawed techniques to their business analysis.) But then, we don't have to IMAGINE, because we've just seen this happen. Right?

    There is NO catch-all system for analysis, nomatter what long & fancy language they use to describe their pipe-dream. If you want a software solution, get a qualified consultant to develop one that fits your unique company culture, niche and business advantage. Don't be first in line to buy a vehicle that "can drive itself."
    matthew_of_cambridge