Understanding Your Audience
Understanding your audience is key to serving their needs. In particular, accept that you are not your customer, and assuming their behavior and motivations are identical to your own is a mistake.
Demographic information cannot be accurately assessed by a Web site - your users are invisible except for their IP address, and even that doesn't tell much except their physical location (approximately)
Web site analytics provide key information about the behavior of users on your site. The author provides some information about the kinds of data that can be gathered and the methods by which it can be analyzed, but his discussion is very superficial and he seems a bit awestruck by mundane things.
He goes a bit into behavioral profiles - Myers-Briggs, and Kiersey. I think his point is that you can "think of" your customers in terms of these archetypes, as a kind of user profile, but he doesn't indicate how you're supposed to know which groups your audience falls into. Seems to me he's expecting you to make assumptions, which is dangerous.
There's a short section (three or four paragraphs) about "user-centered design" in which a designer helps define a task and develops prototypes that are tested by users in a laboratory environment. He lauds it as a good methodology for Web site development, but doesn't say much else on the topic.
He provides a brief description about "personas" and how they can be used in user-centered design - developing a sample personality to represent a collection of customer behaviors and humanizer the statistics. But again, he doesn't get into the "how to" and in so doing provides no useful information (and possibly misleads readers).
He also recommends using a matrix that lists all the possible users and all the possible tasks they might wish to perform. And then he says nothing about what to do with it.
EN: In the end, this chapter provides one useful bit of information - that understanding your audience is important - and provides a survey of some methods for doing that, but much is left without sufficient detail or development to be of use.