Information Age: News, analysis & insight for IT & business leaders

There was a time in the early 1990s when senior executives at major database software companies would readily admit to ‘revenue-shifting’. With the justification that “Wall Street needs to see a smooth curve”, they would move revenue between quarters, contrary to accounting rules, often recognising deals before it was safe to do so. While that artificially enabled companies to meet investor expectations, it did no harm to the interests of executives selling their stock options either.

That was just one illegal practice. Another was when senior sales executives would book contracts with the full knowledge that the customer had been provided with a ‘side letter’, enabling them to renege on the deal at any point. And many did, leaving a big black hole in the accounts that needed covering up.

By the middle of the decade, as the so-called relational database wars (between Oracle, Ingres, Sybase and Informix) peaked, the US securities authorities had grasped the scale of the deception and were hungry for blood. Phil White realised that too late.

He had emerged from relative obscurity to become one of the most celebrated CEOs of the 1990s. With genuine charisma and vision plus innate salesmanship, he transformed Informix from the weakest of the runners in a four horse race to a billion dollar software force that gave the mighty Oracle a run for its money.

Just look at the company’s position going into 1996. Its newly rewritten database, Informix Online Dynamic Server was deemed by analysts to be the best performing and most feature-rich database product around; sales were on a roll, rising 51% from $468 million to $708 million over 1995; the company was highly profitable, with operating margins running at 22%; and cash reserves rose to $262 million. Investors, who saw a 3,400% share price rise in five years, loved it.

There were some well-executed strategies behind that success. The company had hitched its fortunes early on to Unix and was able to capitalise on the mid-1990s corporate enthusiasm for the open standards operating system. Unlike it rivals, Informix had a sound third-party channel network providing a steady stream of recurring revenue. And White ensured Informix was run on a no-frills basis to ensure the maximum sales dollars were pumped into R&D.

That acumen was recognised. Financial World magazine gave him the coveted CEO of the year award – two years running. And across the IT sector, he was given star billing whenever he spoke publicly.

How White went from those heights to become inmate number 92769-011 at California’s Lompoc prison – and Informix crumbled (to eventually end up being sold to IBM in 2001) – is Steve Martin’s canvas.

Sour turn

As a sales manager at the company between 1991 and 1997, Martin witnessed the rise and fall first hand, and his account goes a long way to clarify many of the misconceptions surrounding the collapse of Informix and White’s career: a series of deals turns sour (because of side-letters); other customers refuse to pay after Informix ships an unusable product, the object/relational Universal Server; White and others go ahead and book some of the non-payments anyway, but the company still misses its financial projections by a mile and records massive losses; criminal investigations follow and the board finds a new CEO.

Seven years later, White was convicted of securities fraud. And Martin argues – with some force – that although White was associated by many observers with the $311 million of transactions that were found to have been wrongly booked, the bulk of these were actually firm deals booked prematurely. “Fraudulent transactions accounted for 3% of annual revenues,” he concludes, with the phoney transactions relating to a handful of contracts in Eastern Europe and Japan.

Legal documents show that White made $8.2 million from the sale of Informix stock over this critical period. But having argued that he was simply acting with too much business enthusiasm, White was eventually handed an astonishing sentence: $10,100 in fines and 60 days in prison. As Martin points out: “It seems the world’s greatest salesman had closed the biggest and best deal of his life.”

His book, however, provides insight without much flare. The story is laid out in strict chronological order, with each chapter running through Informix’s fortunes year by year. Martin also seems to have made little attempt to interview original sources, instead drawing mostly on court filings and reports in newspapers and magazines (including this one).

As such, it comes across as an act of catharsis, portraying how a follower of a charismatic leader felt when faith and trust is broken.

For business leaders – especially those in software – this is still compelling stuff; for those who played a part in that decade, it provides a flashback to just how wild those days were – and reminds them of how features like two-phase commit and row-level locking could make or break a database company.