Practical tips for improving site navigation

Taken from Chapter 9 of Search Analytics For Your Site, published by Rosenfeld Media.

Just as Site Search Analytics can improve your site’s search system, it can also increase findability in other ways. Here, we’ll show you a few approaches to improve your site’s navigation and metadata, based on analysing your query data.

Improving contextual navigation
for specific content types

You’ve got ‘em: important page types that occur again and again, deep within your site. Perhaps they are pages that describe the benefits that your products bring. Or they are bios of each of the high-priced consultants you hire out. Or they are your intranet’s critical policies – the ones that all employees had better follow.

Whatever these pages are, they behave consistently – in fact, they may be built by your CMS using the same template. And they are often critical to your organization’s success, because if these pages don’t work well, you may be in deep trouble.

They also may generate quite a bit of search traffic. For example, User Interface Engineering found that over 50 per cent of search traffic on a major commerce site was generated from the department, gallery, and product pages deep inside the site, noting that “most users clicked on more than three links before they decided to search”. The image below is taken from Spending Quality Time with Your Search Log by Jared Spool from January 6, 2010.

These deep page types clearly deserve your attention, and you can use site search analytics in a few ways to improve them:

• Identify frequently occurring (and valuable) page types
  In Chapter 3 of Search Analytics For Your Site, 'Pattern Analysis', I show you a method for using your queries to identify page types that searchers seem to be requesting. Once you know what page types are common to your site, you can move on to these next tasks.
• Determine what’s missing from each page type
  Assuming your analytics application will enable you to do so, analyse the frequent queries generated by each page type by starting with a few samples of each type. Look over each sample’s queries. Do you see any patterns? You might find that users are looking to move to another, related kind of content; these desire paths might suggest adding contextual navigation links to the page type. Or you might notice that users are requesting that more information be included in the page type (which means you’ve now identified a way to improve the content). Below, you can see examples of AIGA's event pages, and my analysis of the desire paths leading from them. Desire paths are desired forms of navigation that a system currently doesn’t support. You’ll often see them in a park or on a college quadrangle: paths worn into the dirt by constant use, unanticipated by the space’s planners.

• Identify problematic members of a particular page type
  While you’re comparing your page type samples, you might find one that’s an outlier. For example, one of your 15 product description pages (or, in the case of AIGA, one of your event pages) may generate very different types of queries or a much higher volume of queries. Is that particular page broken in some particular way? Is it poorly labelled, titled, or written? Is it missing some critical contextual navigation? Or, conversely, is there something special about that particular product that suggests you reconsider how navigation works for your other product descriptions? Now you have an interesting question to consider, and you can explore it further using a qualitative user research technique.

Each of the event pages I sampled indicated that users were interested in getting more information on each of the page’s respective topics. There are many potential reasons for this: for example, the users weren’t able to make the event in person but were interested
in the topic. Or they visited the page after the event had taken place and wanted to know
about the topic. Or they got to the page from a Google search, and it wasn’t quite what they
needed, so they used the AIGA site search to see if it had what they were looking for.
Whatever the case may be, since these events seem to serve as good landing pages for their topics, AIGA may want to invest in linking to relevant (and more static) content on its site from within each event page.

This article is an edited excerpt from Chapter 9 of Search Analytics For Your Site, published by Rosenfeld Media. You can get 20% off all products published by Rosenfeld Media by entering the discount code NETMAG12 when ordering from the company's website.

Picture of Louis Rosenfeld by Myra Klarman Photography.