Lessons from Childhood: How to search for things that are hard to find

tennisball.jpgIn my teen years in the Dominican Republic, I used to play a game my friends called playball. The game followed more or less the same rules of baseball; we setup bases, hit the ball and ran to score. The main difference was that we played in the middle of the street and we used no bats or pitchers. We threw the ball up into the air ourselves and hit it with our bare hands.

Obviously we didn't use real baseballs, which would have been incredibly painful, but rather bright green tennis balls. It wasn’t uncommon to hit the ball hard enough that it will land in a monte—the scrubby woodland areas nearby. In no time at all the ball would grow dirty enough from our monte homeruns that it would turn a dark green color. You always had to pay close attention to the trajectory in order to find where it landed. We lost many balls over the years.

Why was it so difficult? The dark green balls blended in with the bushes so well that it was very hard for us to tell them apart. The balls had no distinctive features to make them stand out. Later, we used bright orange balls and those were really hard to miss. Search engines face a similar challenge…

The search for distinctive features and the role of SEO

As the Web keeps growing at its rapid pace, the features or quality signals search engines use to distinguish documents are not going to be enough. It doesn't matter how hard they work to improve their algorithms, if the documents they are trying to locate are not easily distinguishable, search engines will fail.

The question then becomes: will search engines be able to scale and increase the number of features necessary to differentiate documents as the Web grows? Or, will they need the help of users and SEOs to make the documents easier to find in the first place?

Peter Morville, author of Ambient Findability, envisions the latter. I agree with him, not only because I'm an SEO and he paints a really bright future for us, but because it makes sense. Think about all the content writers that are not savvy enough to use key terms people search for. How will search engines know to show those pages for the right searches? What about other types of non-information objects, like physical ones, how can we search those?

At the moment, it is clear that the main problem search engines face moving forward is the human factor. We're ambiguous, unpredictable. Personalization doesn't help here. Personalized search assumes that eventually computers will be able to know not only what we are looking for, but also to predict what we will want in the future. If that is even possible, it is definitely a gargantuan task. On the other hand, who better to solve the human problem than we humans? There clearly needs to be cooperation between search engines, content writers and SEOs.

Users generally provide more meaningful and useful classification than computers. In the end, relevance is relative. We can look at Web 2.0 tagging as one example of this. This is the type of outcome that we as SEOs need to embrace. Our work is to make our content and our clients easy to find by the people that seek them.

SEOs and search engines are not at war. In the end, we are all working to help search engines do their job better! It is time to admit this publicly.


2 replies
  1. Jez
    Jez says:

    The idea of predicting what people will search for is interesting, but getting the data is difficult. Gmail would be a good source, however Google stated that they would not analyze data against an individual. I think privacy will be a significant hurdle when analyzing data from personalized search.


Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

Your email address will not be published. Required fields are marked *