Google has been around for 15 years, and while they are continually updating their algorithm we’ve been continually updating the way we search. A popular trend that we showed you five years ago was the increasing length of search queries; as people became more familiar with search engines, they began to know what to search for with less trial-and-error.
For example, in the early days of Google you might have searched the following sequence of queries when searching for a car:
- used car
- used ford car
- used ford mustang
- used ford mustang atlanta
These days, most people jump directly to item #4 because they understand how to get the results they’re after.
Siri and Google Now
With the introduction of high-quality voice input via Siri and Google Now (and even Microsoft’s Cortana), our style of search is changing once again. With voice input via Google Glass, Android Wear and (presumably) Apple’s upcoming iWatch, voice search is becoming increasingly prevalent.
Search like you speak
The words we type into search engines aren’t typically the same as the words we speak. Using the example above, I doubt many of us would voice search for “used ford mustang atlanta”. Instead, we might say something like “find a used mustang near atlanta”. Shanti Shunn at Ecommerce Consulting offered a similar example:
- Text: “new pink golf clubs intermediate skill full set womens”
- Voice: “find me a full set of new pink women’s golf clubs for sale that would be perfect for a intermediate level golfer”
The main difference is that when speaking we tend to insert more filler words such as “of”, “a”, “be”, etc. While search engines can filter those out, they can also give the search engines greater insight into the user’s intent. Search engines can struggle at times to determine the intent of a search, and those additional words may help.
Something that search engines have been working on for years now is a concept called “Latent Semantic Indexing”; using additional words in a query to figure out the meaning of other words.
For example, if I searched “photos of jaguars”, what would you expect to see? Google could show you photos of Jacksonville Jaguar football players, photos of sports cars, or photos of an exotic animal. You were only after one of them, but Google couldn’t tell which one.
Instead, if you were to search “photos of jaguar players” Google would use the word “players” to realize your intent behind the word “jaguars”, and then hopefully show you photos of football players.
Write like a human speaks
For years, Google has encouraged sites to write for humans, not for keywords. While keyword analysis and research is important, the end result should be human-friendly text. As we wrote about last month, Google isn’t changing the rules of search, they’re simply getting better at enforcing them. Having users search with full sentences mean that sites with matching full sentences will reap the benefits.
When writing content for your site, think about how someone might search to find you. FAQ pages have traditionally done well because many people type questions into search engines (“How do I change the background on a WordPress site?”) and pages that feature the question will often rank well.
Our friend Tom Tortorici gave a presentation at our Meetup last year titled “Optimized for Humans“, which encouraged site owners to focus more on humans than on search engines. As Google continues to refine their results, and we shift the way we search, optimizing for humans and optimizing for search engines are slowly becoming a single goal.