Those of us who are professional web developers, but not necessarily professional SEO experts, generally think we know what goes into making our pages work well for search and that we don’t need to stay up on breaking developments. I have been like that, and this SEO project has shown me that important changes have taken place without my paying much attention to them.
Last August Google launched a major update to its search algorithm. Like all good IT companies they give cute names to their releases…Hummingbird in this case. Google has made changes to their algorithm in the past. In fact they are constantly making changes. The SEO industry makes its livelihood on understanding these updates and even posts an ongoing timeline.
Hummingbird was a different sort of release though. It was a significant change to the core algorithm that dramatically changed how queries get interpreted. Hummingbird focuses on the entire query as a single entity, rather than a series of individual key words, and tries to predict the user’s intent. In essence it uses the context of the key words to make predictions about what the user will be searching for beyond the specific terms and returns those sites on the results page and the Knowledge Graph.
For example, before the Hummingbird release a return page for “I need a mechanic to fix my 1997 Chevy pickup” might include links to the Chevrolet corporate site, nation-wide auto-mechanic firms, for-sale listings, pickup accessories, etc. A return after the release would be more likely to return local garages that specialize in General Motors repairs.
This type of contextual analysis means that our search tactics have to evolve as well…but this is actually a good thing. Google is trying to get people to the information that they want. So in order for Google to match our site with the search users it must find that information on our web sites. Indirectly, then, Google is forcing us to write better content…or at least content that our users want.
Microformats first introduced me to semantic markup several years ago. I remember a seminar describing how search engines that could perceive semantic information would be able to pull relevant content to make for a better search result. But at that point the concept of a semantic search was still in the nebulous future. That future is quickly becoming the present, and it has shaken up how we look at search, and thereby how we build our content.
No comments yet.