I mentioned that one of the fun things about the retail business is that you can see for yourself how it works. But that is not entirely true.

Let's travel back in time to the year 2000 (remember when that was in the future? You don't? Oh, shut up), when I was a business reporter working for a newspaper in New Jersey. There was a bagel chain called Einstein Bros. Bagels, and everyone loved them. They executed marvellously: beautiful shops, yummy products, reasonable prices, pleasant customer service. Every Einstein Bros. Bagels in the region had a line out the door and halfway around the block every morning.

And they went into bankruptcy.

WHAT HAPPENED!?! people shrieked.

The answer was something that no retail customer could see: They were paying too much rent. Competition for high-traffic retail spaces is fierce, and Einstein Bros. had decided it was worth paying an arm and a leg for desirable spaces.

It wasn't, but that's not the point. The point is, from a customer perspective, Einstein Bros. did everything right--everything. To the customer, they looked like a thriving business. And they went into bankruptcy anyway (although they did come out of it eventually, so there are some benefits to having customers who love you).

Another vitally-important aspect of retail is inventory management. It's one of those things that customers notice only if it isn't happening: You go to the store for milk, and there's no friggin' milk. God damn it, the kids are screaming for milk, and you just wasted a trip--do these people realize how much gas costs nowadays? Screw it, you're never going back there--you'll go to Other Supermarket instead. Sure, they [cost more, are further away, are less child friendly, have surly staff, don't look as nice], but at least they have milk when you need it!

Inventory management isn't just about getting goods into a customer's hot little hands (although that's vitally important: If your business is exchanging goods for money, you'd better have some goods to exchange). It's about doing so cheaply.

Really super-duper, state-of-the-art inventory management makes it cheaper to get your goods to your customers. Wal-Mart is what it is today because they decided, hey, let's have our suppliers deliver goods to our stores. That way, we don't have to pay for gas or hire drivers! (You think I'm kidding, but I am perfectly serious.)

Do you know who else has fabulous, super-duper, state-of-the-art inventory management? AMAZON.

Why? Because Amazon started out selling books. Can you think of a class of goods that's a bigger pain in the ass from an inventory perspective than books? There's millions of titles! Even a single title can have multiple editions! People get really upset when you give them the wrong book! No one knows what books are going to become best-sellers!

Amazon is what it is today because it was able to leverage that inventory expertise into other goods. They're spending $90 million on a single warehouse, for Christ's sake, and it's not because they're building it out of gold--it's because their inventory systems are automated and computerized and very, very fancy. They are so good at getting random goods to the random people who want them that other retailers let Amazon fill their orders, knowing that Amazon does it better.

If you are trying to open up an on-line store that ships physical goods to people, you are going to have to compete with a company that has spent years learning how to ship goods to people, as well as gazillions of dollars on computer systems to help them manage inventory. You have to compete with a company that builds $90 million warehouses. This is no small task.

If you are trying to open an on-line store that sells e-books to people, that $90 million warehouse that Amazon is building does not matter. Their long years of experience in inventory management does not matter. When it comes to e-books, there is no inventory to manage.

This is why I hold the apparently shocking opinion that e-books do not really play to Amazon's strengths. They play to some of Amazon's strengths: Amazon is the dominant on-line retailer of physical books, so when people think "books" they think "Amazon." In addition, Amazon enables book discovery. If you are an author, uploading your book to Amazon is easy, and they offer a good royalty.

All good things. Amazon is executing its e-book strategy very well. Yay them.

But if they fail to keep executing well? Other brands (like, yes, Barnes & Noble, or let's pretend that some Web entrepreneur buys the Borders name) are associated with books, so if they executed their e-book businesses more-competently, they could be more-serious competition. Web sites like Goodreads and Library Thing exist for no other reason but to enable book discovery--that is all they do at the moment, but that could change. Making it easy for authors to upload books to a Web site does not cost $90 million, even if Google can't do it with $19 billion.

It's all about barriers to entry--is it hard for new players to enter a business, or is it easy? If you are talking about selling physical goods on-line, I would say that Amazon has some great barriers to entry--it would be very hard and very expensive to match their ability to manage inventory. If you're talking about selling digital goods...not so much. Amazon isn't without advantages, to be sure, but it's going to be a constant struggle for them--they can't just throw down $90 million and say, "Beat this."

(By the way, if you're wondering why Goodreads and Amazon can't seem to get along any more, I hope that makes more sense to you now.)