The overriding theme to the book is that anything on your web page that takes more than a fraction of a second of thought is bad. When I worked at the Internal Revenue Service, we were never allowed to post anything that took less than a day of thought.
Sure - the topics in this book are obvious. There's nothing here you couldn't have figured out yourself if you took the time to do so. But that's the point - Krug took the time to assemble these obvious but numerous issues for you, so you don't have to think through all of the potential problems your web site is likely to have. IRONICALLY, THE REAL VALUE OF THIS BOOK IS NOT IN ANY BRILLIANT INSIGHTS. THE VALUE IS THAT YOU ARE FORCED TO STOP AND THINK ABOUT YOUR OWN WEB SITE AS YOU READ. That is, simply by taking the time to drift through this light read, you can't help but to ponder how your own web site suffers from each of Krug's common web page problems. You'll undoubtedly end up making a number of improvements to your own site. Krug's small suggested improvements taken collectively really do end up making a big difference to your site. I made at least ten changes to the web site that hawks my own cheesy book (Web Service and SOA Technologies) based on Krug's very good advice.
Weakness #1 - The book's pace slows down at the end. I can't help to wonder if Krug was a little concerned that he wasn't going to have enough pages. Do we really need an entire page that tells us that some people are naturally less patient than others? But even at his slower pace, there are still many sentences that make you think (errr....even if you're not supposed to).
Weakness #2 - Why didn't Krug design a checklist of issues as the last page? You can't use his table of contents as that checklist since his style is to use titles that don't mean anything without reading each chapter. He needs a summary page like this:
* Is it obvious where you can click and where you cannot? Are there "hotspots" on images that are not obvious?
* If a user were to squint and look at your web page, could they discern what each area on the page was most likely about?
* Does your search capability have confusing pulldowns?
* If a user arrives at any random page on your web site (say, from a search engine), can they figure out what site they are on, what the page name is, what are the major sections of the site, what are the best options on the page, where they are relative to the other pages, and how they can search?
* Do your user's eyes have to leap all over the page in order to figure it out?
* Does any operation ever take more than a few seconds to figure out?
* Does the reader ever have to read through instructions to figure something out? (They won't.)
* Is information organized in a clear, visual hierarchy?
* Do you violate any web page conventions?
* Does your site have excessive images and flashy items on it?
* Are your pages reasonably short? (that is, not too much scrolling required)
* Does the page have any text on it that isn't absolutely necessary? (like this parenthetical note, for example)
* Navigation on your site has to be crystal clear. If the user is "looking for a chainsaw", do they know if they should look in the "tools" section or in "lawn and garden"?
* If the user makes a bad guess when navigating your site, is it easy to recover from the error?
* Do any of your pages look so different from the others that the user might be confused if they've accidentally hyperlinked off your web site?
* When you analyze your site, have you spent the majority of your time thinking only about the higher level pages (rather than the low down, leaf node pages)?
* Does every page have a unique identifying name?
* Is every page name prominent?
* Does the page name ever not match the hyperlink that was used to arrive at that page?
* Have you favored the use of navigation tabs? (Krug is a big proponent of tabs.)
* Does your home page establish the site mission, hierarchy, and search method? Do users immediately know why they should be on your site and not someone else's?
* What items appear "above the fold" on each page? (that is, without having to scroll down)
* Does the site have any current references so users know they are not looking at an old, dead site?
* Does the company have a good, descriptive tag line?
* Is it clear where the user can search, browse, and find the best your company has to offer?
* Are you aware that display space devoted to promoting one item implicitly detracts from other items on the page?
* Have you ever observed a completely new user (with no introduction whatsoever) land on your site?
* Have you made the mistake of doing no user testing at all because comprehensive testing is too expensive?
* Did you perform usability testing very early in web development, like you should?
* Did you make the mistake of giving help to your new test user during your usability test?
* Does your site blatantly omit obvious information about your company because of embarrassment? Does it conceal information like contact phone numbers?
* How quick and easy does your site service its most common request?
* How kind is your site to the vision impaired? What happens if you change the browser font setting to "largest"? Anything?
* Does every image have "alt text"?
You may think you don't need to read the book now, but remember the real value of the book is to force you to stop and think a while about each issue. You can't do that blowing through this list in 10 seconds, especially without the examples in the book.
Overall - Despite some minor weaknesses, the book is great and offers enormous value. Don't Make Me Think has been one of the best sellers in software books since August 2005 and for good reason. Krug's "Don't Make Me Think" really does warrant some thought. If you have a web site, there's no question that you should buy this book.
Author, Web Service and SOA Technologies