The Key to Organizing Web Content is Stickiness

Sticky bun.

The stickiest content rises to the top. Image courtesy of Maggie Smith / FreeDigitalPhotos.net

The most important thing you can do to organize your Web content so that people can find it is to make it sticky. Making it sticky is more important than categorizing it or placing it in a hierarchy or taxonomy. It is even more important than linking your content set effectively. In fact, if you don’t make it sticky, neither of those other things are likely to matter much.

It is easy to think of the Web as simply a vast ocean of content. But if it were that, it would not work at all. What the Web actually is is a vast index of content. It is not a fixed index, like in a book, but a complex, dynamic, volatile, multi-stream index. For purposes of findability, how your content is organized on the Web comes down to how it appears in that index. And while you can definitely contribute to how it is indexed in small ways, its indexing is largely controlled by others. The Web organizes itself communally.

That index is made up of the indexes of all the search engines, but also of all the links on Twitter, Facebook, LinkedIn, and other social sites, and all the links in other content across the web. It includes all the references in all the forums, and every mention in a private or group email. Every like, every thumbs up, every +1, is a part of the index. Every time one entity on the Web points to another, it is part of the index.

You may also think of your content as static and permanent, and therefore having stable organization. But because its real organization is contained in the index that is the Web, and that index is volatile and controlled by others. The organization of your content is changing all the time, even if the content itself isn’t changing.

To put it another way, you don’t organize your content on the Web. You put your content on the Web and the Web organizes it. You can’t force how the Web organizes your content, but you can influence it. Your first concerns, therefore, should be to create content that the Web will organize in a way that is favorable to your ends. In other words, your first concern should be to make your content sticky.

Sticky content is content that sticks to other content. It is content people link to and select from search results. The stickier your content is, the more prominent it will be in the great index that is the Web.

Stickiness is not a general property, however. If it were, the entire Web would be one big sticky mess. Rather, stickiness is highly dependent on circumstances and audience. What sticks to one thing is repelled by another. Being sticky is all about being sticky to the right things. That is how your content will find its audience, by sticking to that audience and to the things that audience likes.

This business of stickiness is not unique to the Web. Stickiness was important in the age of paper as well. Unsticky manuscripts were repelled by the publisher’s slush pile; sticky manuscripts ended up on the best seller shelf at the front of the bookstore. The best way to make sure that people can find your novel or non-fiction book is to get it on that coveted shelf space by the door, and that is only done my making it sticky.

The Web is not merely the media, it is the marketplace. In the paper world, there were many gatekeepers along the way who tried to judge the potential stickiness of things before they were eventually submitted to the ultimate stickiness test of the marketplace. Many things that would have been sticky if given the chance, never saw the light of day, because the gatekeepers never saw their potential, or because their stickiness only appealed to a market too small or too distributed to serve profitably.

On the Web, though intermediaries do exist in some places, by and large the marketplace is open and direct. Put something genuinely sticky on the Web and people will find it. This has been clearly shown in the growing sales of indie music and obscure books that were never commercially viable in a world of atoms, but find their audience effectively in the digital world. The Long Tail exists because the Web organizes content based on its stickiness.

How do we make our content more sticky? Much of that is specific to your subject and to your audience, of course. But there are a couple of basics that should be universal.

  • Make bite sized pieces. Content that the reader has to chew on is not as sticky as content that is easy to consume. Write Every Page is Page One topics, not books. 
  • Focus your energy on stickiness. Yes, relationships and classification also play a role, particularly in facilitating local navigation within your content, but they are irrelevant unless your content is sticky enough to draw an audience in the first place.

, , , , , ,

8 Responses to The Key to Organizing Web Content is Stickiness

  1. Cassie Buchner 2013/04/22 at 11:43 #

    Any more suggestions on how to make our content more sticky? I appreciate the suggestions you provided, but I’m thinking there has got to be more than just two.

    • Mark Baker 2013/04/22 at 16:47 #

      Hi Cassie

      Thanks for the comment. Indeed, there are more. Many, as I mentioned, are specific to the subject matter and the audience. Good content is sticky content, and how you make content good depends a lot on who it is for and what it is about.

      But I will have some more things to say about making content sticky in future posts.

  2. Tom Johnson 2013/04/22 at 16:14 #

    Good post. I think you make a great point here. Important, relevant content naturally bubbles up to the top of the index through all the links, comments, and other web-karma that adds to its stickiness.

    I wouldn’t discount the value of marketing, though. Great content needs a little prompt for visibility sometimes.

    As a related question, do you know how to surface the top 10 most important articles on a website? For example, let’s say that for new users visiting a blog, we want to dynamically surface the top 10 most popular articles on that site.

    I think marketers do this regularly when they place text links. Certainly the scope would come into play here. Top 10 results based on X keyword. Maybe google just does that automatically based on their Pagerank algorithm, but which keyword searches would you use? It would be neat to see this kind of thing in action.

    • Mark Baker 2013/04/22 at 17:05 #

      Hi Tom. Thanks for the comment.

      Indeed, marketing can certainly help. What is interesting is how marketing works on the Web. It can be very hard to push things in a pull medium. It is much more about getting things to be pulled, which kind of takes us full circle to stickiness. You can certainly do things to seed the index, though — like tweeting about blog posts, for instance, and hoping people retweet (and don’t unfollow you).

      Don’t you have a widget on your blog that lists the most popular posts based on the comments they have received?

      Of course, the idea of surfacing the most popular things on a site works on the presumption that people come to the site directly, rather than Googling for information. If they Google for information, they are much more likely to land on one of the top 10 most popular articles than to land on the list of the top ten articles.

      Listing the most popular things probably works for a site like Amazon, where people tend to go to Amazon first and then search that site for what they are looking for. But if you are not a destination site like Amazon, listing your best stuff on your home page may not do you much good because that listing is only a teeny tiny part of the great index that people are using to find content on the subjects you cover.

      The other thing about listing the 10 most important articles is, most important to whom? Before you log in, Amazon shows you the best sellers. As soon as you log in, it shows you things it recommends for you based on your profile.

      What you really want is to show each visitor the 10 most items on your site that are most important to them. But the fact of the matter is that the Web is much better at doing that than you are. And people are much more likely to trust whatever mechanisms on the Web they trust than they are to trust your site.

      So, if your visitors arrive not on your home page, but on the page that the Web things is the most important one for them, your opportunity to keep them on your site lies in that page. This is why I think links are so important. Links are part of the index of the web, and the part that you can control totally within your own site. Provide great and plentiful links within your site so that the reader has less incentive to wander off to Google again.

      To look at it another way: the thing about the index that is the Web is that it is not a central index. It is a distributed index. There are little bits of it on every page the reader looks at. So the trick really isn’t to gather up all your indexing in one place, but to distribute it around your site, the way the Web distributes its index around the Web.

      • Tom Johnson 2013/04/23 at 14:33 #

        I forgot about the Most Popular widget on my blog. I actually dislike the automated popular posts widgets because they never surface what I want to surface. I have a post with a picture of a grasshopper that seems to always top the charts, but that’s not what I want users to find.

        You’re right on with your observation about relevancy being determined by the search. And I agree that most search engine automate this process anyway.

        I did just add a couple of related links to my sidebar. If you look at a post now, a Related Links widget shows post titles and excerpts from the 10 latest posts. If you arrive at the site via a search engine (e.g., search for “technical writing careers”), another Related Posts from Search Engines widget shows related posts based on search engine keywords.

        I think both of these related posts widgets will be much more relevant to users who arrive at a site via search.

        I love that I could implement both of these features in 5 minutes (searching for “Related Posts Widget”. from the Plugins > Add New screen). If I were authoring with a help authoring tool rather than using a web platform, I’m not even sure how I would begin to implement these features except by an elaborate system of tagging (rather than keywords). This is one reason why i think web platforms offer so much more capability with help material. When we stick with the old HATs, we really trap ourselves in the past.

        I think I once wrote about HATs versus web tools here.

        • Mark Baker 2013/04/23 at 15:04 #

          Hi Tom,

          I agree with you completely on HATs vs Web platforms. And I will suggest that the difference between them comes down to whether they architected for book-like content or for EPPO. HATs are book-like. Web platforms are (generally) EPPO.

          The reason this makes such a difference to their flexibility and extensibility is that HATs have a threaded content architecture and Web platforms have an unthreaded content architecture. With a threaded architecture, there is only one principle of organization and that is built right into the tool, and thus right into the content as it is authored.

          With an unthreaded content architecture, organization is not build into the system or the content, just various pieces of metadata that can be used to organize the content after the face. This opens up the field for all kinds of plug ins that either create and manage metadata, and/or exploit metadata to organize content in new ways.

          Book-like structure assumes that organization is fixed at authoring time. EPPO structure assume that organization is dynamic at read time. Everything else flows from that.

  3. Virginia Lynch 2013/04/23 at 21:34 #

    This is a great post. And I particularly like the sticky bun because it says it all. And when you’re looking at that sticky bun, you get pulled into it like you do on a website that is organized so your eyes and your intuitive senses provide the directiion and focus. Thanks again.

  4. Vinish Garg 2013/04/23 at 22:16 #

    Mark, I enjoyed this post particularly for the way you lend focus on how web content organization is different from content in print. I feel that making content sticky is less of an information architecture challenge and more of a challenge for what help authoring tools we use. And this is where I agree with Tom that HATs are giving place to web platforms.
    I have used WordPress, core HTML (with bit of PHP), and we are even planning to use Drupal for developing online manuals simple because of scalability and flexibility (for findability, better user-experience, custom theming such as for top ten topics, integration with 3rd party helpdesk systems, social media integration, and so on) to customize the help systems. During the TCWorld India Conference in Feb this year, I had a discussion with Mike H on how Flare plans to evolve for being ‘social’, They have taken the first step in latest release but the noteworthy changes are planned for early 2014 and so I expect that HATs will continue to struggle to match the power and pace of evolution of web platforms.