Increase visitor stickiness by site design?

A WebProWorld Post asks "Are there characteristics of the design itself that help make a website more sticky?"

Assuming that there is content of value to the visitors (not always a valid assumption), then yes, site design can improve stickiness.

For example: most of my traffic comes from Google search and when it is very specific, I get no stickiness. Someone searching for "Lost Linux Password" finds my article, reads it and they are gone - because obviously they have an immediate problem and the article solved it. Maybe they'll come back later, but if they've found their answer, now isn't the time to be noodling around the site: they are off to do whatever they wanted to do before the forgotten password issue stopped them.

The searcher who got misdirected by Google isn't likely to stick around either, especially if their search is way off: a musician searching for guitar strings gets sent to an article here that mentions "string theory" - that person is gone before the page finishes loading. But even a not quite satisfactory search usually won't result in browsing around: most of us return immediately to the search results page when a particular page doesn't pan out. The answer we seek may be on another page right there, but if it is, we'll let Google find it for us rather than use the site's own search tools.

A regular reader isn't as likely to browse around either - they came to read today's post, and they already have read the one from yesterday. They might want to poke around a little sometimes, but most of the time its one page and out.

Nothing in the way of design will change the behavior of those people.

But there are people on more generic quests. They come with keywords of "Linux" or "Unix" or by following a link from some other site. I have a much better chance of keeping them around and getting them to read a few more posts. And yet..

Until I changed my design I wasn't doing well at that.

In fact, I was running very close to one page per visit, period. Given the tremendous influence of a number of specific purpose problem solving articles here, I never expect to have a very large "average pageviews" figure, but I knew it could be better than it was, so I started making some changes. I use Alexa to track how well I'm doing; that's a little rough because you have to wait a few days to see if your changes are working, but where I was at a flat 1.1 average page views, I'm creeping up toward 3 now and expect it to keep improving. What changes made the difference?

One was devoting more space in the sidebar to related pages. In 2006 I had "Related Pages" at the bottom of each post but had taken it out because it was difficult to maintain and I couldn't devote much space to listings. You can see the results in the Alexa graphs: 2006 had good average pageviews but that plunged back down in 2007. People like related content and are very apt to click on one of the pages offered.

Another are is the index pages by topic. Every post is tagged, and every post offers to display more that has the same tag. The problem was that the page that took them too wasn't very conducive to digging around: it was just a boring list. What I changed recently was to grab teaser text from articles and randomly display three of of them at the top of the listing. That's also helped considerably.