IT Strategy: Predict able

David Rae

Hurricane Katrina, terrorist activity and struggling public companies don’t
have a great deal in common. They do, however, form a central plank of the sales
pitch for SPSS – a software company whose CFO I had coffee with the other day.

As pitches go, it was rather impressive. Their software was used by the US
Centre for Disease Control to predict where illnesses would be likely to hit in
the aftermath of Hurricane Katrina. Together with another software provider,
SAS, their technology is also used to try to predict where terrorists will next
strike. And text-mining technology the company has developed has an important
place in the armoury of many investment bankers – the program searches through
US public companies’ Form 10-K filings looking for evidence of whether the
company should be classed as a buy or a sell. (The story goes that certain
phraseology used in such filings signifies a company in trouble. If the word
‘strategic’ appears liberally in a company’s financial report, for example,
investors should run for the hills.)

SPSS operates not in the extremely fashionable world of business intelligence
(and, for the record, I’m a fan of business intelligence) but in the even more
niche world of predictive analytics. In other words, it not only presents you
with a digestible view of the business, but, in theory, it recommends what you
should do with the business as a consequence of that data. (Watch out, if you
haven’t already realised, the machines are taking over…)

SPSS has been around for a long time (it was apparently one of the first
technology companies to list on the London Stock Exchange), but it has always
operated just beyond the public’s range of vision. Its main competitor, SAS,
however, has been increasingly visible of late – its sponsorship of the Six
Nations Rugby and consequent advertising campaign (including several within
these pages) are both evidence of a company that sees serious growth potential
and a maturing market just around the corner.

But having access to such technology, I can’t help wondering whether SPSS and
SAS could have turned their view-finders onto their own industry. Because while
we were supping coffee at a predictably posh hotel in central London, technology
giant Oracle was putting the finer points on its $3.3bn cash acquisition of
business intelligence company, Hyperion.

The Oracle deal caught a few people off guard. Quite apart from the fact that
the deal takes Oracle’s spending spree to $20bn in just three years, the
acquisition is a direct attack on its German nemesis, SAP. While SAP has
remained predictably quiet about the acquisition, it doesn’t take a genius to
work out that it will be quietly and, no doubt, efficiently seething behind
closed doors.

Hyperion, as Oracle president Charles Phillips gleefully pointed out to the
world in a press release announcing the deal, is used by thousands of SAP
customers as their financial consolidation, analysis and reporting system. “Now
Oracle’s Hyperion software will be the lens through which SAP’s most important
customers view and analyse their underlying SAP ERP data,” said Philips. Ouch.
Oracle never was one to back away from a scrap.

It’s likely that the Oracle announcement will spark the latest wave of
consolidation in the software industry. The pure-play business intelligence, or,
depending on who you talk to, business performance management, sector comprises
companies such as Hyperion, Business Objects, Cognos and Cartesis. All partner
with the big three software companies (SAP, Oracle and Microsoft) and have grown
to become significant businesses in their own right. And we all know that when
businesses grow up, Oracle CEO Larry Ellison tends to buy them.

Now that Oracle has swallowed Hyperion, how long until SAP takes a bite at
either Cognos or Business Objects? And how long then until Microsoft hoovers up
the remaining player? Or even gets in sooner? All of which leaves Cartesis – a
former PricewaterhouseCoopers’ company which was sold to private equity group
Apax Partners in 2003 when it became clear that Sarbanes-Oxley was impacting on
its potential to grow while within the PwC stable.

Which all helps to explain the rather jittery email that landed in my inbox
about 35 seconds after the Oracle/Hyperion announcement was made: “This is good
news for Cartesis,” said Crispin Read, the chief marketing officer of Cartesis.
“However, we are concerned for Hyperion’s customers.” Which seemed remarkably
compassionate, considering the circumstances.

The statement went on to list Cartesis’s concerns about which Hyperion
product lines will be continued by Oracle and how quickly the rate of churn of
Hyperion customers to Microsoft and Cartesis products will rise.

Maybe Cartesis is right and Oracle will discontinue some of Hyperion’s
product lines, leading to the odd disgruntled customer. But to me, the sounds
coming from Cartesis remind me of the sort of noises a threatened animal might
make upon realising that its natural, and blissfully safe, habitat had just been
invaded by lions. I wonder if they predicted that.