I’ve been giving a lot of thought to bundling lately.

Generally, I think bundling is a pretty stupid idea. Bundles don’t exist to solve problems for consumers. They exist to solve problems for producers, either subsidizing products that can’t be sold unbundled (cable packages), to maintain manufacturing volume (magazine subscriptions), disguise the real cost of a product (Tivo, cell phones), lock out competitors (Internet Explorer), or to create the illusion of value where none exists (local telephone services beyond dialtone and call waiting).

Because bundles seldom solve problems for customers, they generally fail. Even when they continue to survive in the market, it’s not clear that if the components were sold individually, the net present value to the producer would not have been greater.

I can only think of a few situations where bundles make economic sense. One is where individual components of the bundle could not be sold economically, but that there is enough value in the bundle to satisfy customers. Norton Utilities is an example of this kind of bundling. So is Microsoft Office. Back in the seventies, this could be said of cable programming, but now cable programming could be unbundled if cable companies weren’t monopolies.

A second example is volume sales. Magazine subscriptions are a good example. Publishers cut out distribution middlemen and increase readership. Readers are given an almost impossible-to-resist bargain as a result.

A third is where you need to create a bundle to solve a chicken & egg problem or, in rare cases, to create a whole that is more than the sum of its parts. Apple Computer is a good example where the combination of hardware and software is genuinely synergistic. The early cell phone industry might be another example where bundling hardware and services might have made sense, but technology has passed this model a long time ago.

The fourth economically successful kind of bundling is the forced buy. Generally, you have to be a monopolist to force customers to buy a product that they don’t want to get something that they do, or give away a product to force a paid competitor out of business. “B” movies and a lot of pre-antitrust IBM software were subsidized in this fashion. Microsoft used this technique (as well as an excellent product) to make Internet Explorer the standard Web browser. Generally, these bullying tactics can be successful, but are bad for end users and markets, and are often illegal.

Outside of monopoly markets, I don’t believe it’s possible to create a successful bundle that doesn’t create real value for the end user by either reducing price or creating something genuinely new.

If you can’t do that, you should be asking what your real motivation for bundling is.