~ OT ~ A VERY INTERESTING post from the @home thread...
<< Sorry, I must disagree. I acknowledge that ATHM has a higher price to sales ratio, which was presented in my last post. But that doesn't make it more valuable. On the most straightforward measure of market capitalization, CSCO is about 10 times as valuable as ATHM. But the math error isn't what troubles me about your statement.
What troubles me about ATHM and its monstrous valuation is that it is just a transactions company. Transaction services don't have the market dominating characteristics to make them 800 pound gorillas. And without those characteristics, this company can never begin to justify its current market capitalization.
If ATHM had both proprietary architecture and high customer switching costs, the case could be made that it will grow into its current valuation. But it doesn't have both and as long as DSL and AOL are available, switching will be relatively easy for ATHM's customers.
The reason I selected CSCO for the comparison in my last post is because it is an 800 pound gorilla and it has proprietary architecture (on routers) and high customer switching cost.
The following excerpt is from page 126 of Geoffrey A. Moore's book “Gorilla Game,” which came out about one year ago. Its discusses the merits of investing in transaction services companies and why long-term they can't become gorillas:
“Transaction services, as opposed to professional services, are bought not as one-time projects paid for on a fee basis, but instead are consumed recurrently and are typically paid for by subscription or a pay-as-you-go charge. Telephone calls, FedEx packages, airline tickets, dry cleaning, and movies are all sold as transaction services. In high tech, the Internet is consumed as a transaction service, with Internet access being the initial payment transaction, and e-mail and Web browsing being the primary usage transactions to date. But even product companies, once they get to Main Street, will make an increasing portion of their revenues from transaction services in the form of product maintenance, technical support, outsourcing, and the like.
Transaction services offer a particularly attractive business model because once the system's infrastructure is built out, and once revenues exceed fixed costs, the variable costs are so low that virtually every incremental bit of revenue drops straight to the bottom line. Market-leading transaction services franchises such as the SABRE System and AT&T Long Distance, therefore, are some of the most profitable enterprises on the planet. So why, then, aren't UUNet, Yahoo, or AOL part of the gorilla game?
First of all, the two conditions for getting these enterprises profitable are extremely hard to meet, especially during a tornado market. Building out systems infrastructure is hugely expensive, as business plans for any of these companies make painfully clear....Historically, most successful transaction services franchises come into being on the back of publicly funded, not privately funded infrastructure. The original Internet itself represents such a publicly funded entity. But now the commercialization and scaling up of the Internet is being funded privately, at potentially ruinous costs. It is not clear that this model can ever pay its investors back. So that is strike one.
Strike two is getting the revenues. Where will they come from? Subscription fees, we have already discovered, are not enough. Consumers and business customers alike are too price-sensitive. For a while we thought transaction-processing fees would be the source, but the volume of transactions that one could charge a fee for is still very low in the Internet space. Maybe someday. Charging for information is a third possibility, and maybe someday it too will come through, but for now there is so much free information on the Internet that paid-for info is being shunted aside. All this clears the way for the current knight on the white horse-advertising. Internet sites, most business plans now assume, will act like TV channels-they will attract viewers, and these viewers' eyeballs can be sold to advertisers.
Let's assume that this model stabilizes. (Please note, however, that it hasn't yet, so investors should garner a higher risk-adjusted premium for going forward.) Now the problem is to get the costs under the revenues. Again, there are all kinds of ideas about how this might be done, but little track record to date to show that it has been done. Indeed, the press repeatedly calls attention to the fact that the best known sites are as yet unprofitable. Moreover, there are whole industries, like the airline industry, that appear to be constitutionally unprofitable. There is no guarantee, in other words, that this model will ever work. So that is strike two.
Strike three, for us, is actually a slider, which may or may not have hit the outside corner. (This is where you really need an umpire.) It has to do with our core issue: proprietary architecture and high switching costs. In a transaction services business, the analog to proprietary is brand, and the analog to switching cost is community. The TV shows Seinfeld, Friends, and Frazier are brands. So are Yahoo and AOL. The TV shows are also communities-they have definable demographics, and moreover, they create word-of-mouth communities as people talk to each other about the episodes. Yahoo and AOL thus have the potential to provide not only the topics of conversation but also the medium-be it a chat room or a threaded discussion group.
So let's assume that the winning sites achieve this kind of status. (Please note, however, that this too has yet to come to pass, so investors should garner a second upward adjustment in their risk premium.) Now the question is, what are the switching costs? On TV, they are not very great. That's why the ratings shift around so much. Why will they be higher on the Internet? Does anybody think they would even approximate the switching costs of replacing even a modestly successful product architecture?
The answer is no. Therefore the notion that these companies could be gorillas simply does not hold. The risks they entail are much greater, and the security of their leadership positions, once achieved, is more precarious. They must, therefore, be some other kind of animal-python might be a good choice. Once again, we are not saying that pythons are not worthy of investment (although in this case, the danger may be higher than usual). Nor is it to say that their stocks will not appreciate dramatically in the next few years. They may beat every other stock on the market, for all we know. But if they do, it will not be because of the dynamics of the gorilla game. Some other dynamic will have to explain it. The transaction services sector is not a gorilla habitat.>>
-------------------------------------------------------------------------------------------------------------
Uncle Frank: I tend to agree with the author. IMO, there are VERY HIGH switching costs that develop once a customer starts experiencing the FULL BENEFITS of a Direct Relationship with DELL.
You may have a new recruit here for your "Gorilla and Kings Thread." <gg> .!!
Best Regards,
Scott |
