I have glimpsed the future of search, and it is not keyword-driven. While I have long been an advocate of using keywords as an indicator of a searcher’s intent, I am about to eat my words.
The truth is that search is heading in a direction that most of us could not have foreseen… a technically complex and varied amalgamation of platforms, devices, and inputs.
What I mean by this is that as the search engines become more focused on discovering user intent based on various elements that can be measured before a user even types anything — location, search history, mobility, circles, etc. — it becomes less important what that searcher actually typed in to access the SERP. And with the rise of conversational search touted in platforms like Google Now and Google Glass, the searcher may not even type at all.
Take a recent example of how Google Now works. You can ask, “Who is the President of the United States?” And the answer is displayed for you: it’s Barack Obama. You can then ask, “Who is his wife?” The answer is again displayed for you — it’s Michelle Obama — but let’s say you choose to click on or otherwise select a result on that page. Maybe it’s a listing of famous first wives throughout history. As the site owner, the referring keyword would be [who is his wife]. That’s not useful to you, because you don’t know if [his] refers to Obama, Jefferson, or Washington.
This is the future of keyword-based referral, and it’s one reason of many that you shouldn’t be too upset about “Not Provided” going to 100%. (Unless, of course, you want to debate the politics of the issue, data sharing, and paying for data — then there’s plenty to get upset about.) But on the face of it, you’re not losing much in the way of useful customer data.
No, SEO is not dead. And neither are keywords. The truth is, we probably won’t see the standard search box disappear in our lifetime. But we will see a lot more variation in how people type keywords into the search box.
These changes make it absolutely essential to grow SEO into a slightly different concept. Gaming the system ended last year, for those of you who didn’t get the memo. Attempting to reverse-engineer the algorithm will get you in a lot of trouble.
But the future of SEO is all about optimization. In a sense, what’s old is new again — except that the way we think about the term [SEO] needs to be changed again. Matt Cutts said at SXSW last year that we should think of SEO as “Search Experience Optimization.” I’ll go one step further and suggest…
Instead of thinking of “marketing,” which is defined as the act of promoting and selling products or services, SEOs need to be thinking about how they can deliver the best possible experience for their subject — the visitor.
What combination of elements do they need to present to make the user experience most optimal? What key actions do people want to take on the website, and how can we appeal to their base instincts and language with clear headlines and copy?
There’s a concept called “aboutness” that was developed by R. A. Fairthorne back in 1969, popularized by William John Hutchins in the mid-70s and more recently adopted by Shari Thurow, an industry expert whom you may be familiar with. Originally used for library and information science, in a marketing application, it refers to making what a page is “about” very clear to the user.
You can do that with well-chosen and labeled images, with keyword-based headlines, and with copy that clearly explains the purpose of the page. This is where keywords become essential, because until we can read minds, we have to be able to guess at the language that will compel users to take action — guess at first, and test to refine. This is not to be confused with “user experience optimization,” which goes deeper and is more detailed. But it does scratch the surface of what good SEOs should be thinking about.
The traditional definition of SEO, “Search Engine Optimization,” will always be relevant and necessary. This refers to how we optimize websites for search engines, and it includes everything from making sure search engine spiders can crawl pages to helping them understand complex content on pages with Schema markup. But there’s another, even more important, aspect of SEO that we will need to pay attention to.
Earlier this month at SMX East, there was a panel (the first of its kind, I believe) on Entity Search and its impact on the future of SEO. David Amerland began his part of the session by asking everyone to think of a tree. I immediately visualized something like the image on the left (below), but then realized he probably meant for people to visualize something like the image on the right:
The rest of this exercise was particularly powerful for me because of what I had visualized at first. David said you probably saw something different based on where in the world you live, or what your understanding of a tree was. In my case, my mind was on entities and information architecture, which is why my “tree” looked like the image on the left.
The power of this example is that it explains a universal truth in a very concrete way: language cannot indicate with certainty what someone is thinking.
If Google had known based on my location or the last photo I took or my search history that I was working on information architecture for a client during the last break or attending a search conference, then a search for [tree] might have resulted in something very different for me.
So as SEOs under my new definition (Subject Experience Optimization), we have a responsibility to clearly define and indicate our clients’ entities. Where are they located, what do they specialize in? How do those things form relationships to other entities? Some of this is familiar, like the concept of “local search” or “authorship” or “link graph;” but, the whole is truly greater than the sum of its parts.
The future of SEO is not based on keywords, but rather on how those “keywords” form a relationship to an entity, a concept, or a target. For more on this, I strongly suggest reading Paul Bruemmer’s latest article on Entity Search.
As to what exact format that will take, or whether we’ll all soon be flocking to FreeBase in the way that we once killed DMOZ, I can’t say. But the future of search is coming quickly, and SEOs (whatever we define that as) will need to adapt rapidly to remain ahead of the pack.