September 19, 2002

37signals should know better

37signals generally does awesome stuff, but I just noticed that their home page has a nasty little behavior: if you click anywhere inside one of the three columns on their home page, it is a link to another page. What's bad is there's nothing to indicate the body text is clickable, and my mouse cursor doesn't indicate it's a link. Anyone clicking from one browser window to another might click on that text, only to be suprised that it's a link. That's how I happened to notice it in the first place. Come on, guys. Links should be obviously different from body text -- I have to believe anyone that creates something as good as Design Not Found knows this...so why the goof?

Just realized this is my second critique of the 37signals home page...
Wireless as marketing gimmick

[W]e built a WAP add-on for our MLS product to allow Realtors to search for properties over their phone. Not one of our customers ever bought the WAP module, but it was still something we needed to have available. Most RFPs we would get from potential clients requested that wireless access be available. So we built the module, not because we thought that anyone would ever buy it, but because it helped us win sales. Why wouldn’t people buy it? WAP sounds cool when you describe it, but in practice, it’s more difficult to use than it’s worth.
Giving the Human Touch to Software

Making too many assumptions about users’ expectations and levels of competence can get software developers into a lot of trouble. Yogita Sahoo tells her own story about designing an application for an industry she was deeply familiar with—but that industry knowledge didn’t keep her from making some big usability blunders.

All the terms used in the above message looked very obvious and simple to my team and me. We took for granted that the rest of the world also knows what a file menu is and that clicking on the top “X” button will close a dialog box. But unfortunately, the hotel employees didn’t know about these conventions. Had we understood that a steward would not be familiar with computer terminology, the product could have been designed to suit a layman’s needs.

You should never try to design for a wholly indeterminate set of users. Your marketing team may add some insight, and a human-factors specialist will also help. You should work with a representative user group that varies in terms of profession, age, and qualifications.

September 18, 2002

Site Refactoring

Kalsey Consulting tells us how they implemented the concept of refactoring to improve their site. They cover card sorting as a way to recategorize and restructure, how to move files without breaking things for users, and generally making improvements in an evolutionary manner.

Refactoring is the process of making small changes to a program that improve the overall execution without introducing new features (and hopefully no new bugs). The basic idea is to leave things better than you found them.

[N]ow I had a solid reason for moving my files around. Placing everything into the new structure would make it easier to manage in the future. I made the decision to move everything into the new structure, but also decided to minimize the problems created by doing so.
Book Excerpt - eSupport
E-Support : How Cisco Systems' Saves Millions While Improving Customer Support

The most important finding was that we were letting "feature-creep" get in the way of usability. We added in lots of bells and whistles, like sorting and document rating, but these were either not noticed or criticized for getting in the way of the task at hand. Our next iteration will have a simpler interface. Because we had not invested any time developing the systems to support these bells and whistles, we could easily discard those ideas without wasting precious development time. (from page 2)

More posts related to features and usability tradeoffs
- User centered design sells products
- Bloatware: Good or Evil?
- The Pursuit of Simplicity
Open Letter to a Power User / Developer

I just read this Letter to a Non-believer, and have to respond. As I see it:

1) Someone commented in a mailing list that Linux has "poor usability."
2) You point out that *you* can successfully use Linux to read email, write professionally (about Linux and technology it appears), compose music, watch movies, plan Linux events, create your own Linux distribution, publish Linux CDs, and browse the web, etc.
3) You claim that "millions" of other people who "work like you do - productively and happily" also use Linux.
4) Since you can do all these things, you then assert that Linux *must* be usable. You say "you have the gall to tell me and millions of others that it can't be done"?

Well, I DO have the gall to tell you your logic is horribly flawed.
1) Usability is relative - something that is usable for one type of user doing one type of task is very often not be usable for all user doing any kind of task. Linux is obviously at least somewhat usable for some folks, but that doesn't meant that it's usable for most people.
2) Usability is not black and white - it's not "usable and unusable" it's shades of gray. At some point individuals determine that things aren't "usable" enough for them - this is the point where people either buy into or pass on something.
3) Usability is only one small factor in the adoption of products.
4) You are obviously a power user of technology and Linux. When you say you maintain your own distribution you reveal that you are a power user among power users. Few "average Joes" use Linux for average tasks, and my guess is that few actually could. (Notice I said few, not none.) I know many Linux users, but noone that maintains their own distro.
5) Let's talk realistic stats - where are the millions of folks in regular offices or homes doing average mundane things on Linux? How many non-programmers run Linux? Sure, millions of servers run Linux, but that's not what we're talking about. I'm sure there are millions of Linux distributions sold every year - that doesn't mean millions are actively running or anywhere within reach of a "typical" consumer.
6) Okay, even if Linux with KDE or whatever were super easy to learn and use. Where would a soccer mom buy a preschooler edutainment for Linux? Could she install it and read the docs (don't get me started on man pages)? How about a tax package for my small business? Can I get it at Best Buy? Power users have different needs and understand how to locate Open Source needles in the haystack of the Internet - average folks want quick, easy and mainstream. It's not just the OS that has to be usable and suitable- OS's are just the start. It's the whole offering from the platform and all the related software vendors. Why do you think Apple still has any market share? It's because they have enough of the right stuff (usability, software, marketing, documentation, service, etc.) that people want when making a decision on what to adopt.

Just because YOU use it, doesn't mean it's USABLE for many people. Show me research - maybe an independent usability test. Lots of universities use Linux -- ask them to research Linux's usability. As far as adoption statistics are concerned, back up your cited numbers, show me real numbers from reputable analysts, All I could find are gross estimates from biased partisans.

Linux rocks as a server platform, and it's a great development platform for many developers (depending on what they develop in and for). Linux on the average consumer's desktop? Not in the foreseeable future - it's built by geeks for geeks. And geeks love it, so it's successful in its own way.

September 16, 2002

Shopping carts aren't just for purchasing

Shopping Cart Abandonment: Why You Need More of It points out that online cart abandonment rates will always be higher than the 2-3% in physical retail stores.

According to a white paper by Fry Multimedia, "Most (online shoppers) appear to use the cart to mark products of interest, like turning down or marking a page in a catalog. Items in shopping carts on Web sites represent shoppers' desire to purchase, not necessarily their intent."

Because of this, it's unlikely that online abandonment rates will ever be comparable to traditional ones. Users have a hard time finding what they want on the Internet; shopping carts provide an easy way to bookmark things that they are interested in.

Abandoned shopping carts aren't just shopping carts. They also are important sources of customer information. How important? Imagine what it would cost to put together a list of what each of your online customers was interested in buying!