IBM supers shun nukes for biz analytics
BAO. It's the next ERP
As part of its briefings to Wall Street and investors this week, IBM trotted out its recently announced System S streaming system and the "Watson" QA super. Both are variants of its BlueGene/P Power-Linux supercomputer lineup, but both are aimed at solving business problems instead of simulating weather or nuclear explosions.
Both machines are examples the underlying hardware that Big Blue expects to be the foundation of a huge wave of spending on business analytics and optimization, which the company's top brass pitched as the next big opportunity for hardware and software sales and which they said would grow to rival sales driven by the enterprise resource planning (ERP) software boom that started in the mid-1990s.
This BAO opportunity is quite large. (Heaven help us. We don't need another abbreviation but we also can't say business analytics and optimization every bloody time either). And just as ERP did, it cuts across all of IBM's divisions and groups. Speaking earlier this week at IBM's investor day, Frank Kern, general manager of IBM's Global Business Services group, said that the BAO market was just an evolution that has been taking place for decades as companies automated bookkeeping, moved on to MRP-II software to manage their manufacturing operations, and then extended out into automating the whole enterprise, from shop floor to front door, including supply chains and customer relationships, with the ERP phase.
(Kern's Global Business Services group doesn't do systems integration, outsourcing, maintenance, or other kinds of tech services. It raked in $19.6bn in 2008 selling more slippery kinds of services such as business process re-engineering and optimization).
Alongside of ERP, companies have deployed data warehouses and data marts to mine their ERP data to trends to help them better sell products and run their businesses in a more lean fashion. Business analytics and optimization wants to take this one step closer and actually create systems that can predict outcomes for business decisions before those decisions are made.
Such capability, which IBM's nerds at its Research division as well as in its Software Group and its Systems and Technology Group, is most likely one of the reasons why Big Blue can be so confident that it can hit its profit numbers in 2009 and 2010, which it was bragging about at the investor day conference. IBM must be running scenarios constantly to help it shift people around the globe to chase opportunities.
Kern said that the market opportunity for BAO products and services was about $105bn in 2008, which is 18.6 percent of the much wider "business automation" opportunity, which includes all manner of software, hardware, and services sold to corporations and governments to run their operations, which Big Blue pegs at around $566bn in 2008. Now, here are the bits that have IBM salivating about this BAO segment. Between 2007 and 2012, the BAO segment of the IT industry will have a compound annual growth rate of 8 per cent, which is twice the growth rate over that term for the industry as a whole.
And within Big Blue itself, which is obviously the heavy hitter in IT services, BAO is expected to grow at around 10 per cent this year and accelerate to between 15 and 20 per cent from 2010 onwards, accounting for more than $2bn in sales in 2010. Within five years, IBM expects BAO to drive as much sales for it as core ERP systems do.
IBM has caught the bug for predictive business decision-making as much as it ever caught the buzz for relational databases back in the late 1980s, data warehousing back in the 1990s, Linux in the early 2000s, and cloud computing today. You'll have to judge for yourself how much of this BAO stuff is real and how much is vaporware or putting a new paint job on data warehousing.