Tuesday, May 10, 2005

Part II: What Would Grandma Think?


Patrick Byrne likes to say he runs Overstock.com not for the satisfaction of the Wall Street “jackanapes” who nit-pick his company’s income statement and question whether his growth now, profits later mantra is good business or just good spin.

Instead, he talks about things around Overstock passing something he’s called “the grandma test”—a sort of What Would Grandma Think? standard of honesty and decency when it comes to reporting the facts behind Overstock.com’s numbers.

He particularly dislikes the notion that “widows-and-orphans” may, when it comes to stock market fluctuations, be at the mercy of “Wall Street criminals” (Byrne’s term), and saves his most potent venom for hedge funds:

“They inject noise into attempts to have adult conversations with our shareholders. And the thought that any widow-and-orphan money gets invested based on the work of such hedge fund lickspittles makes me nauseous.”

If anything should make Patrick Byrne nauseous, however, he might look to the persistent inability of his company to hit targets he has set publicly in shareholder letters and on conference calls—targets for new web offerings (travel, build-your-own-ring, auctions), new ventures (mCommerce), and new technology (“Project Propeller”).

While we explored some of these missed targets in “The Mystery of the 38 Diamonds,” yesterday in Part I of “Patrick and the Amazing Technicolor Dreamcoat” we began looking deeper into the largest Overstock.com mystery of all: whether the company's growth now/profits later mantra is in fact good business strategy, or merely good spin.

However, since I run a hedge fund, let’s not look at this from the point of view of a lickspittling jackanapes like me. Let’s look at it from the point of view of the “widows-and-orphans” for whom Patrick Byrne expresses such tenderness.

Let’s ask “What Would Grandma Think?”

Well, for starters, just last month Overstock shocked Grandma by announcing a Q1 loss of $4.2 million. Grandma was shocked because only three months before, Patrick Byrne had announced a $2.5 million Q4 profit, using words like “superb,” “brilliant,” and “flawless” in his January letter to Grandma and other Overstock investors.

In that letter, Byrne had written to Grandma: “For several quarters my letters and conference calls have described at length our internal work building an Ark: the waters came, and our ark floated magnificently.”

He ended it by telling Grandma that Overstock was “up on a plane from which I believe only our own inattention or mistakes will knock us for some time.”

So as Grandma reads Patrick's April letter, she is naturally wondering what, in the space of three months, caused the “ark” to spring a $4.2 million leak. The letter starts with “eight items of interest,” including the incomprehensible-to-Grandma “large binomial marketing experiment,” by which Byrne explains much of the loss.

But Grandma smells a rat. She reads the letter carefully and concludes that at the heart of the reversion to red ink is a sudden, huge increase in technology-related spending. She is, of course, confused, because she remembers how “the ark floated magnificently” only three months before. So Grandma decides to go back and thoroughly comb the record of Patrick Byrne’s letters to shareholders, as well as the conference calls.
.
And here’s some of what she's finding.

Overstock’s Information Technology Department

Grandma always liked reading about Shawn Schwegman, the “star” of the Overstock technology team. Patrick talked about him so glowingly:

Q2 2004 Letter: Shawn Schwegman (VP, Technology) remains a star. This year Shawn has assembled a team of mature, experienced Database Administrators (DBA's). Our network has become quite stable… By the end of July, I think I will be able to say, "I could not be happier with the IT team nor the level of understanding between it and its internal customers."

Q3 2004 Letter: I could not be happier with the IT team nor the level of understanding between it and its internal customers. I have never worked in a company where the president could say that about the IT department. This took Shawn Schwegman one year to achieve.

Q4 2004 Letter: Shawn Schwegman (VP, Information Technology) and his team saw us through Q4 with an almost spotless record. In addition, Shawn has arrived at a wonderful vision of how our systems can be modularized to support growth with flexibility for years to come. There will be quite a bit of internal development and re-architecting over the next six to nine months, but some changes are necessary, and at our present size we believe the payoff can be enormous in terms of customer experience and cost savings.


So far, so good: Shawn was doing wonderful things with Overstock’s IT department. He had this modularized vision that could support growth “for years to come.” Yes—there would be “some change,” but nothing alarming, it seemed.
.
Now, imagine the shock Grandma gets reading the transcript of the April 2005 conference call:

Q1 2005 Call: Shawn has a term—decrapitate. And our internal systems, to be honest, are decrapitating. That’s why we’re switching to this other. We’ve got people nursing the systems along here, the systems we used to manage marketing campaigns and things like that are decrapitating, in Shawn’s word.

How, Grandma asks herself, did Patrick Byrne go from telling me “our network has become quite stable” and promising “growth with flexibility for years to come”…to all of a sudden telling the Wall Street jackanapes “Our internal systems, to be honest, are decrapitating.”?

Decrapitating?
.
Grandma scratches her wig. This “nursing the systems along” doesn’t make sense. Patrick had written so many times about “leading edge” equipment and “robust” systems. She digs out more old letters and conference call transcripts, and finds Patrick's discussion of Overstock's technology systems.

Overstock's Information Technology Systems
.
Grandma quickly finds what she is looking for:

Q2 2003 Call: We have dramatically expanded our capacity in the last three months…. And our servers—we’re moving to a clustered Linux environment. We are actually just days away from turning on—switching to a clustered Linux environment. We had outgrown HP N-class and rather than go to a large Sun or an IBM Superdome or something…


And again! Right here, Patrick says his IT systems are cheap and scaleable!

Q2 2004 Call: … As far as IT systems are concerned…I cannot say enough good things about it now. I know I sound like a commercial for Oracle, but it truly has given us remarkably cheap, scaleable computing power.

Then Grandma reads the transcript of the April 2005 conference call. It doesn’t make sense:

And here we’re tying to build a business that does a billion and 2 billion…we were stretching things very thin…. So this early February we just decided to pull the trigger. We had gotten everything pretty well architected and we just said we’re going to make these changes now.

It’s just that we’ve grown past the stage that we can do things with Excel spreadsheets and hand calculators


“Excel spreadsheets?” Grandma mutters. “Hand calculators? What about those large Sun or IBM Superdomes he was talking about?”

She recalls something about Oracle, that big database company run by that Larry Ellison-fellow. She was always so impressed when Patrick talked about his big Oracle system.
.
Overstock's Oracle Database System
.
She finds the first mention of it:


Q3 03 Call: We threw the switch on a project…an Oracle database cluster. I am given to understand that there are only about 10 companies who are successfully operating on this platform and only five that have built as large a cluster as we have. That just went live a few weeks ago.

Only five companies! Imagine that!:

Q1 2004 Call: We're doing things somewhere between the leading edge and the bleeding edge I'm afraid to say. For example, the Oracle cluster. We know of other companies that are trying to get their cluster live. We've been live with it for six months. Oracle is showing us a lot of love and they've actually been a very good partner with us in building this out.

Grandma knew that Ellison fellow was a decent man. She reads on...:

Q2 2004 Letter: Last year our database server reached its limits: we migrated to an Oracle cluster solution that is now highly (and cheaply) scaleable. We have outgrown our data storage solution, so we are buying a robust SAN (Storage Area Network) that should carry us to several times our current size. … As far as IT systems are concerned…I cannot say enough good things about it now. I know I sound like a commercial for Oracle, but it truly has given us remarkably cheap, scaleable computing power.


It all sounds so wonderful, but Grandma's hands are sweating and her wig is slipping off: after the let-down with that nice Shawn fellow, and all that IBM Superdome stuff, she is getting nervous about what might come next. And sure enough:

Q1 2005 Letter: A good way to think of it is this: we had a huge database with all our product information, customer information, even logistics and financial information, contained in it. That database grew into an enormous ball of spaghetti. It became difficult to manage the computational "bowl" needed to hold that ball, and the situation is getting unstable. Our answer is to untangle the spaghetti into smaller bowls, each of which is itself a cluster (so we end with clusters of clusters).

Grandma is shaking. From “leading edge” to “an enormous ball of spaghetti”…in only one year!
.
In the manner of a person who buys a car that is, in fact, a lemon, and has suffered the first few minor headaches with growing nervousness, Grandma is beginning to wonder if she's getting the whole story on Overstock.com.
.
She is beginning to question whether its CEO does indeed, as he writes, set the gold standard in communicating with candor the results of her company.
.
Tomorrow we'll look over Grandma's shoulder as she digs deeper into the question of what really is happening behind the scenes at Overstock.com.
.
Jeff Matthews
I Am Not Making This Up
.
The content contained in this blog represents the opinions of Mr. Matthews. Mr. Matthews also acts as an advisor and clients advised by Mr. Matthews may hold either long or short positions in securities of various companies discussed in the blog based upon Mr. Matthews' recommendations.

7 comments:

hedged said...

doesn't patrick byrne's brother run a hedge fund that berkshire hathaway seeded? i wonder if patrick is happy with having jackanapes in his family

k9thunder said...

Well, more contradictions or white lies from Byrne. No suprises really. Byrne living and beliving in his own little lies and distortion sheltered from reality in a fantasy world busy fighting the evils of many kind who frankly are the reasons of OSTK recent downfall. My recommendation is that he just ought to shut and try to run this peddler of unwanted stuff. It's money losing retailer!

Keep up the good work Jeff and thanks so much for revealing the FACTS!

smr said...

>>a lickspittling jackanape like me<<

Suddenly I feel like I'm in grammar hell . . .

The word jackanapes is singular. Jackanape is not a word. Lickspittle is a noun. There is no verb to lickspittle.

BelowTheCrowd said...

I thought there would be no reason for me to return to the discussion of OSTK, as I neither have a position, nor follow the company.

But as a guy who has spent the bulk of his career making strategic IT decisions, I find it hard not to comment.

The first and most obvious conclusion, which probably can be made by a non-techie as well, is quite simple: He has an IT department that likes expensive toys regardless of need and without any semblence of a coherent strategy or architecture.

Not the place you want to be if you're depending on the stuff to run an ongoing business.

Also, obviously they like bleeding edge stuff. Now, I hardly oppose an occasional project of that nature, but you don't do it with your mission critical systems. You make small experiments to familiarize yourself with the technology, in places where it won't demand lots of money and won't kill you if it fails.

This is the kind of thing for which a solid management team is a godsend. A visionary CEO rarely had the kind of operational knowledge he needs to rein in this kind of stuff. The dotcom era is littered with "visionaries" who allowed their underlings to spend the company into bankruptcy. A strong, experienced COO type is usually what's needed to hold back the natural tendency of all techie types to spend money on toys, and to convince the CEO that he shouldn't be so impressed with the technological marvels the IT guys keep presenting him with.

-btc

cdub said...

This blog is Overstocked with information on Overstock. Jeff, there are other stories, companies, shady CEOs, etc. out there...

mfairview said...

Looks like people are getting the idea about Jeff's alterior motives wrt his (and his buddie's) OSTK shorts. Monday's meltdown and rumors of some Hedgefunds (perhaps Jeff or Jeff's buddies involved?) having to liquidate due to Kervorkian's gambit on GM. Yet we hear nothing from HedgeHog Matthews.

Direwolf said...

Hi Jeff. Steve B. from over on Street Insight. Rob M. told me about your blog today so I stopped by to check it out. Great job. I try to do something sort of similar for my clients. I'd love to get some insights from you. Get my info from Rob and pop me a note or IM. Thanks.