Google’s search engine parses through millions of web pages and still delivers amazing search results.
Still, even Google’s search algorithm has its limits. With more and more web content available every day, you can see why Google offered additional filters to help you narrow your search results. For example, if you’re looking for pictures of pumpkins, you can type “pumpkins” in the search field and click Images. Alternatively, if you’re interested in the latest books about Thanksgiving, you can search for that term and click Books.
Companies, such as Dell, have used this technique—called faceted search—for some time. For example, on Dell’s web site you can search for a laptop or desktop computer. Alternatively, you can use the filters to help you search by product category, processor, screen size, weight, and so on.
For users, this provides a lot of benefits including the ability to be much more specific when looking for certain types of content.
Faceted Search and Technical Communication
There has not been a lot of discussion about faceted search and technical communication. I believe that if your help system is large (for example, more than 500 topics), a faceted search could help your users search by topic type (such as troubleshooting content, procedures, and conceptual material), training vs. technical communication content, HTML versus PDF guides, and so on.
There are, however, a few problems with using faceted search with the current Help Authoring Tools.
First, most Help Authoring Tools don’t offer faceted search. So unless you’re using a tool that does offer the functionality, you won’t be able to provide it to your users. It’s worth mentioning that MadCap Flare does offer a form of faceted search in its product.
Second, in my experience, the search in Help Authoring Tools isn’t great. For many years, Google’s search engine has set the standard for finding content online. In contrast, the out-of-the box search found in a Help Authoring Tool is mediocre.
Third, I find it’s hard to customize the search engine in a Help Authoring Tool. RoboHelp’s search does allow you to tweak content but the last time I looked into it, the search engine is largely a black box that isn’t meant to be significantly customized.
What do you think? Would faceted search help your users? Should it be it available in all Help Authoring Tools?
Mark Baker says
Google’s search engine is so powerful because it is able to do statistical profiling of billions of search strings and terabytes of content. Search engines for desktop help can’t match Google’s performance because they don’t have enough data to work with. This is part of why the best place to find a needle is now in a haystack.
A faceted search is actually more of a query than a search. Queries work well when the user understands both the categories and the values in those categories, as is generally the case in shopping sites like Dell, but not generally the case in technical writing.
More on needles in haystacks: http://everypageispageone.com/2011/10/12/the-best-place-to-find-a-needle-is-a-haystack/
More on the difference between search and query.
Tom Johnson also has some worthwhile stuff on faceted search: http://idratherbewriting.com/tag/faceted-search/