Courtesy of Susan Hall of IT Business Edge

At a time when it's doing massive cost-cutting and layoffs in other parts of the business, Procter & Gamble CIO Filippo Passerini plans to quadruple the number of people on staff with expertise in business analytics, InformationWeek reports.

The article doesn't put numbers on how many people that will be, but it's following a trend. A survey a few months ago by Accenture SAS Analytics Group found that 72 percent of respondents plan to increase spending on business analytics this year, yet 60 percent said they don't have the needed skills in place to effectively use analytics.

InformationWeek's Chris Murphy explains that P&G is switching from preparing reports for key people after the fact to creating a virtual, real-time war room of people huddled around video screens. Videoconferencing plays a big role in this, with players worldwide able to join in even from a smartphone.

The article says P&G isn't looking for new data types for these meetings - it's still relying on point-of-sale, inventory, ad spending and shipment data. But it's looking for more frequent updates faster and with a finer level of granularity. It's moving beyond what a problem is and why it's occurring to bringing analytics in to solve problems faster. Writes Murphy:

• Passerini pictures analytics experts ... sitting in on more meetings to make sure the "how" to solve problems gets sorted out right then and there, not postponed until everyone gets more information. The old model would mean "let's get back to this in two weeks," he says. "You need to be able to answer that question immediately."
   

A commenter on the article sees problems with the classic Dilbert model of "get as much data as possible so that the executives can ignore it and make decisions based on company politics." Yet Murphy explains that rather than trying to assemble by guessing which data executives want, Passerini wants to make all the most recent data available so there's no questions about whose data to use and what other information is out there. The analysts sitting in on these meetings - a breed in short supply - will be business-IT hybrids.

A second commenter sees the need for more predictive analytics, writing:

• People used to get reports which told them what happened, but not why. They now know what and why, but the cutting edge step is moving to understanding what will happen and why it will happen using modeling tools before they implement a decision.
   
• Many organizations, if we are honest, do not even know to a high level of certainty what happened. They get BI reports and everyone immediately questions the validity of the numbers. Master data management and data quality is a necessary first step for many organizations before they even are serious about BI or analytics.

In a second, really interesting article on P&G's efforts to increase transparency and agility, Michael Schrage at Harvard Business Review writes that greater transparency in large organizations generally brings on more micromanaging. At a conference, he asked the P&G execs how they dealt with that. He was surprised at the answer: have fewer managers, with the remaining managers too busy to meddle.