Overstock.com (OSTK): Part Five –The Thinking Part (Competitive Advantage & Risk)

There is something almost self-hypnotizing about running scenarios through highly-tuned valuation spreadsheets. Just a tweak to this variable, a rub on that number, a slight nudge here and hit calculate…

Behold! I have wrought a thing of beauty! It is my child. I am proud of it. It can do no wrong.

It is easy to convince ourselves that these models can accurately distill an entire business – in all its complexity – into a simple sensitivity matrix. This represents all possible variations of the future, we tell our beaming selves, as we put a particularly thick border around the cells containing the middle-option valuation. The middle-option, not too bold and not too conservative, is always the right one.

Yes, we’ve uncovered some impressive ROIC numbers in this business. Yes, we’ve seen how it shouldn’t be too wild an assumption to see earnings grow near or above that $45 million threshold we set. And yes, we’ve supplied this nice, conservative 15x multiple that even a loan underwriter might stamp with his approval.

But now we must step away from our precious model and actually think. We must ask ourselves, are we really considering the right things here? Have we adequately weighed the big risks that face Overstock.com and, by association, our investment? How are we protected?

I wake often in the middle of the night. A two year old daughter is not without culpability. Last night was the norm, and as I attempted to settle back into sleep my brain began mulling through the big risks to Overstock’s business.

I was thinking to myself, the internet sure is a great place to do business. Just consider, you can become a $1 billion retailer without ever opening a storefront. You never fight over real estate in power shopping centers with the most wealthy patrons. You don’t have to hire all those expensive employees. You don’t have that pesky shrinkage problem. Distribution is so much easier. You build a relatively inexpensive ecommerce system. You cobble together a network of suppliers. You run a few Super Bowl ads. And you just watch the money role in.

It sounds so easy, anyone could do it.

Wait! That’s not good. That doesn’t say much for barriers to entry. If anyone could do it for very little invested capital, how defensible is Overstock’s business?  That reminds me of an industry article detailing Amazon.com’s use of fulfillment partners…and Buy.com…and now Sears.com…and (gulp) Wal-Mart.com. They all like the idea of selling merchandise on the web without ever paying for the inventory.

I’m recalling now how Overstock’s business tanked from 2005-2007. Customer demand fell off a cliff because of a disastrous ERP implementation. If the debt and equity markets had not allowed Overstock to raise more cash, it would have been lights out. While that cloud had a nice lining – it did lead management to change the inventory model and significantly reduce capital needs – it still speaks to the tenuous hold the company has on customers. You just don’t get much margin of error…if you don’t execute with precision, shoppers go elsewhere and it’s very, very hard to get them back.

Despite the better expense structure, less invested capital, and a couple years of net earnings, can we really say with confidence that another snafu wouldn’t send customers bolting, revenue falling, and losses mounting?

On top of that, could I really feel any confidence in the valuations scenarios I ran based on the previous two years of operating results? If this is a competitive industry, two years doesn’t tell you much if you don’t have that moat. And as I’m now in a more critical frame of mind, look at those variables…they’re all over the place! The expense structure ranges from 2 to 11% annual growth. The direct business revenue shrunk 13% one year (though it would seem as part of a concerted effort to move investment burden over to their partners for certain categories). And the fulfillment business has a pretty wide range of sales growth that probably can’t be estimated conservatively by averaging out one year of 21% and one year of 10% growth.

I’m wide awake now.

So, what’s the competitive advantage anyway?

Is it Overstock’s relationship with the fulfillment partners? The company depends on them peddling their wares on its site and not going somewhere else. What is Overstock providing that makes them happy? Some thoughts: 1. A marketplace of buyers; 2. $60 million of annual marketing to keep traffic up; 3. IT systems to manage their catalogs, selling, credit cards, performance statistics, etc.; 4. Customer support services so they don’t have to do that part; and 5. Management of all product returns.  There’s a lot of value there, right?

Okay, how could any of the big players start taking that away? Let’s play devil’s advocate and pick on Amazon.com.

For beginners, it could create better terms for the partners. If Amazon offers comparable services and decides to give its outlet partners a bigger piece of the action (hypothetically charging 10% instead of 12% fees on all purchase transactions), that might entice sellers to at least consider jumping ship. And looking at Overstock’s margin needs (i.e., the gross profit they must get from fulfillment partners to cover operating expenses), it doesn’t look as if they could afford a protracted battle with Amazon that depresses gross profits. Amazon, with its $9 billion in cash, has considerable more staying power than Overstock.

Second, Overstock depends heavily on Google search engine optimization and keywords purchases. Management suggests a big chunk of its marketing budget goes there, and Q1 2011 results took a hit when Google put Overstock in its “penalty box” for SEO rule infractions. (Indeed, Patrick Byrne estimated that two months in the penalty box, where Overstock results were excluded from regular Google searches, cost the company 4 to 5% of its quarterly sales. That was the difference between break-even and turning a profit.)

This suggests to me an unhealthy dependence that competitors could exploit. My own experience shows that products sold by Amazon tend to get ranked very high in natural Google searches. And it’s very rare that Amazon products don’t go to the top of the paid search results. The company spends a lot on search words, and it can afford to do it. So, if Amazon puts Overstock in its sights, it begins carrying more of the same products and throws more resources at ensuring those products are search engine optimized and/or at the top of the paid list. Amazon could, without breaking a sweat, outspend Overstock many times over to ensure best Google placement.

Third, if Overstock and Amazon are selling the same or similar products, and Amazon wants to target Overstock to take away market share, the company with deeper pockets has no compunction with starting a price war. I cannot see that Overstock can keep customers without offering an equal or better price.

Now I recognize that this list of risks might just highlight my lack of knowledge about the nature of Overstock’s business. These doubts are, I admit, very likely given my unfamiliarity with the dynamics of the surplus goods market. Perhaps Overstock (for reasons I haven’t been able to imagine in this exercise) has such a dominant position as the buyer/partner of preference among surplus goods sellers that it has a sustainable competitive advantage that Amazon can’t reasonably penetrate. (Overstock’s 19% gross margin on fulfillment partner business suggests it charges its partners higher commissions than either Amazon or the others. That would point to some degree of loyalty from its partners, since they don’t jump ship, which could be a competitive advantage. Up to this point, however, I haven’t been able to confirm this as accurate.)

And perhaps Overstock.com has created such loyalty among its customer base that it wouldn’t fragment even for slightly better pricing. (Q1 2011 numbers show new customers down 15% over the same quarter in 2010, yet revenue was flat, and gross margins were up 6%. It does look like existing customers are repeat buyers.)

But in my research, I was unable to connect with any Overstock fulfillment partners to get a sense of their relationship with the company. Nor do I think the recent customer data is sufficiently long in the tooth to be bankable. (If readers understand the business better than I do, please share your insights! I’m not wed to my risk conclusions here if facts or better interpretations prove me wrong!)

Finally, I fear there are few constraints in the world of internet retailing that prevent a single player from achieving near ubiquity. It truly could be a winner-take-all kind of market. Unlike in traditional retailing where competition plays out in matches of real estate chess, distribution strategies, and the ability to satisfy local preferences, a single internet retailer could dominate by becoming the marketplace of choice for nearly all products and for nearly all buyers. This is especially true if you take away the cost of inventory (a working capital constraint to growth) because fulfillment partners are carrying that burden for you.

It certainly mirrors my own experience as a shopper. Amazon is my default. If I want to buy something, I search for it there. Like most of the country, I’ve been buying from Amazon for 15 or so years. I like the user experience. I’m comfortable with the interface. I trust its security measures. And I’m confident that I’ll get timely delivery of my orders. It would take something very compelling to lure me away.

Charlie Munger, vice chairman of Berkshire Hathaway, notes the competitive strength scale, brand and familiarity can bring to business with the following example:

“If I go to some remote place, I may see Wrigley chewing gum alongside Glotz’s chewing gum. Well, I know that Wrigley is a satisfactory product, whereas I don’t know anything about Glotz’s. So if one is 40 cents and the other is 30 cents, am I going to take something I don’t know and put it in my mouth, which is a pretty personal place after all, for a lousy dime?”

If Amazon achieves this status, I don’t want to compete with it unless I have a distinct advantage that protects me in case the giant wakes up and decides he doesn’t want me around anymore. Maybe I’m too small for him to even care – indeed, maybe the giant would just prefer to buy me, brushing off the gadfly in the process – but I’m not sure I’m comfortable putting real money behind that bet.

With Overstock, I’ll confess that I do SUSPECT that an advantage exists. But the ability to confirm that advantage falls too far outside of my circle of competence to turn a suspicion into a conviction.