Trying to unravel the bundling fallacy

By David Rohde
Network World, 11/01/99

By now it's a familiar scene: The CEOs of the two merging carriers stand up and justify their actions by noting that customers "want to get all their services bundled by a single carrier."

Then the financial analysts weigh in by concluding that whatever its other challenges, the merger offers users "the ability to buy bundled services from one provider."

Then a regulator steps forward, makes noises about the possible anti-competitive impact of the merger, but concedes it "could help make bundled services for end users a reality."

In the megamerger game, the idea that users are looking for bundled services is universally accepted. There's only one problem: It's not true.

OK, let me revise that. Sure, all other things being equal, customers would like to get all their services from one provider on a single bill. But nobody buys something as dicey as a communications link based simply on the fact that the same vendor sold the customer another totally dissimilar type of service.

As a consumer, I'm not going to choose my cable TV vendor - if I ever get a choice - based on which company happens to provide my longdistance. The choice is going to be based on programming options and the realistic possibility of fast Internet access. Period.

Likewise, I'm not going to stick with a wireless carrier with gaping holes in its geographic coverage just because the same parent corporation happens to provide flat-rate Internet access or is running digital subscriber line ads this month. Who would?

On the enterprise side, the whole history of the Internet argues against the idea of bundling. If it hadn't been for independent ISPs, nobody would be using the Internet or even IP browser-based intranets as a platform for enterprise communications today.

To be sure, there are specific subsets of bundling that have appeal. Nobody likes finger-pointing between a local and long-distance carrier. So it's better to get an end-to-end connection from one carrier - if the carrier that provides it is not disorganized from its last merger.

Likewise, lots of users find it convenient for the carrier to provide and manage the customer premises equipment for a WAN. But that's no argument for the carrier to buy the equipment vendor.

Isn't it interesting how the people who always talk about bundling are the vendors, not the users? You would be amazed at how completely unquestioned the bundling idea is in Washington and Wall Street.

Early in my career at Network World, I remember an internal discussion along the lines of "what do users want" when our editorial director, John Gallant, remarked: "I think what users want are products that work."

This amazingly simple and yet almost totally ignored observation has stuck with me ever since. In the real world of networks, people want the best connections at the lowest prices with the finest support. That's it. And if it means the accounting department has to write two or three checks each month instead of one, so be it.