Google recently implemented their new “Hummingbird” organic search algorithm, perhaps the company’s most significant overhaul in more than a decade. Thomas Claburn at Information Week explains that Hummingbird is an expansion of Google’s Knowledge Graph, which was
“introduced last year as a way to help its search engine understand the relationships between concepts rather than simply matching keywords in documents. The Knowledge Graph structures data, so that a search for, say, Marie Curie, returns facts about her contributions to science, her life, her family and other related information, not all of which are necessarily contained in the same document.”
Danny Sullivan’s FAQ at Search Engine Land noted the drive toward understanding how people actually articulate what they’re interested in (especially as we move beyond text-based searches with other interfaces, such as voice search and Google Glass). Searchers will continue to move from keywords to conversations.
Danny used an example search: “what’s the closest place to buy the iPhone 5s to my home?” Rather than simply find pages that include the words “iPhone 5″ and “buy,” Danny notes that with Hummingbird, Google can better understand that you’re looking for a physical store (the “place”) near your current location that carries iPhones (and that an iPhone is an electronic device).
Until recently Google’s infrastructure didn’t have the ability to understand the entire query rather than individual word. If you revisit the video Google published last year of an internal search quality review meeting, you’ll see that part of the discussion centered around the fact that for longer queries Google’s algorithms could only evaluate portions at a time, rather than the whole.
Those of you with a longing for deeper technical considerations will probably appreciate Bill Slawski’s analysis at SEO by the Sea.
For me, in my new career in the search industry, this update comes at an interesting time. I’ve long advocated for greater customer focus and integration across audience touchpoints, and while this seems a simple and obvious enough goal, obstacles abound (from new and evolving communications technologies and channels to traditional internal political dynamics). For too many organizations, search remains on the periphery, unintegrated into the strategic fabric of marketing and communication development. Search pros are frequently relegated to the role of tactical specialist, brought in after the fact to lipstick the pig.
Viewed from this perspective, Hummingbird represents an incredible opportunity.
Google’s basic philosophy seems pretty simple: help searchers quickly find exactly what they’re looking for. On the Web publishing side of the equation, that means they work to mitigate spamminess and gaming tactics, penalizing sites that fail to answer user questions and provide meaningful, useful content and elevating those of high quality and utility. On the searcher interface side, the problem is considerably more complicated. How can the zillions of mind-numbingly complex ways in which a human brain frames a query be translated into a workable mathematical language?
- First the searcher has to understand what he or she wants. Sometimes this is easy – who delivers pizza in West Seattle? – and others not so much – I want something different for dinner tonight.
- Then the searcher has to figure out what word or sequence of words, in what order, best articulates the query. There’s occasionally some effort involved in translating the desire into Google’s language, and most of us have probably encountered cases where we had to try several times before we found the right term.
- The search engine has to interpret the request and route it to the pages that best address it.
- Finally, the searcher has to hope that the Web site with the right answer/information is presenting it in such a way that the appropriate page pops up in the search results, preferably at very top.
Each of these steps embodies potential signal:noise challenges, and at the core of things is an essential technological reality: Google speaks a little bit of human and humans have learned to speak a bit of Google, but at present the two do not have a shared fluency in any language. Each is getting better, but the conversation isn’t yet as fluid as we’d all like it to be.
Back to Hummingbird and what it tells us about the future. The hypothetical perfect search experience blows right through Semantic Web 3.0 and on into what, for lack of a better term, I’ll call the Intuitive Web – Web X.0. In this idealized, sci-fi world search is an almost telepathic process. The user wants X, types/speaks the query, and up comes exactly what he/she asked for. Feel free to think of this in terms of Amit Singhal’s famous Star Trek story.
Larry Page once described the perfect search engine as understanding exactly what you mean and giving you back exactly what you want. It’s very much like the computer I dreamt about as a child growing up in India, glued to our black-and-white TV for every episode of Star Trek. I imagined a future where a Starship computer would be able to answer any question I might ask, instantly. Today, we’re closer to that dream than I ever thought possible during my working life. (More here.)
The results transcend the limitations of articulation (the system noise associated with the user finding the right words), addressing what was meant. They may even include something that’s better than what the searcher was envisioning – truly intuitive search is going to include a healthy serendipity function.
An example. Say I’m a little fuzzy about that night in Kauai a few years ago and I’m trying to tell my friend about this incredible drink I had. I know it was some sort of rum punch and we were at this fantastic steak place. That much I remember. Intuitive Google could quickly sift through all of my confusion and figure out that I’m probably talking about the Tai Chi at Brennecke’s Broiler. (Which, btw, I can’t recommend highly enough.)
In other words, thanks to Hummingbird and its progeny, Google has learned to speak human. If we’re deep enough into a perfectly integrated big-data/CRM world the intuitive/predictive potential for Google is staggering, and I’ll leave it to you to decide if you’d like to be thrilled or, having seen Minority Report, terrified.
After reading numerous articles about Hummingbird over the last couple of weeks, I think I’m beginning to see the general shape of Google’s grand vision. They want to migrate from our current model, where search is a process comprising several discrete steps, each rife with static, into something that’s so responsive it’s completely transparent. From the user perspective, this is a really attractive proposition.
Search companies, though, are faced with the task of evolving, preferably in a strategic fashion. While technical optimization is explicitly about minimizing noise and maximizing signal, the content side has some unfortunate history re: spamminess – link farming, lack of relevant content on the page, overlapping content, packing pages with keywords, etc. These sorts of manipulative tactics are pure noise generation. In an ideal, organic world, there’s a theoretical one-to-one link between what I want and what your company provides. As soon as you and your competitors begin gaming the process, though, you have introduced artificiality into the process. The goal isn’t to hook me up with what I need, it’s to hook me up to your bottom line. If I ask who you think the best band in town is, I don’t need some guy I don’t know (who’s being paid by Citizen Dick’s management agency) “optimizing” your answer, right?
Given this, you’re probably just as happy as I am when Google smacks down the spammers.
With Hummingbird (and Penguin and Panda and 100% [not provided]) Google is moving inexorably in the direction of perfecting the signal:noise ratio. In a world where its algorithms are increasingly intuitive and predictive, there will be less and less call for ad hoc, tactical SEO as we currently know it. The task will no longer be about optimizing content after the fact – it will be about producing helpful content that tells a client’s story in the most useful manner for the searcher. Google X.0 wants content that’s compelling and honest from the get-go. As the graphic in Eric Enge’s piece at Search Engine Land put it, Google wants you to act online like search engines don’t exist.
I’m a storyteller, and there have been plenty of times when bad SEO marching orders got in the way of how I instinctively wanted to describe the value of the service or product I was describing. I was encouraged to communicate poorly, frankly, because the system had to be manipulated. And this manipulation necessarily imposed barriers to the free flow of communication between myself and my audience.
Google X.0? Tell your story, clearly articulating your offering’s benefits and value. Act like search doesn’t exist.
In other words, Hummingbird is great news for everybody. It will be wonderful for those of us searching – you won’t have to figure out what words Google will understand because now it speaks your language fluently. It will be great for businesses – instead of worrying about how to game their sites to deal with keyword tyranny they can simply get back to clearly describing their features and benefits. It will be great for other (that is, non-commercial) Web publishers (like S&R and your favorite political, music, sports or hobby blog) because intuitive search levels the field by favoring content that provides the audience with information it cares about.
It will even be good for search firms, although yeah, the game is changing for them. This is the core idea that Vanessa Fox was articulating, way back in 2010, in Marketing in the Age of Google. In a nutshell, the search industry needs to be envisioning its own future in terms of integration into the broad marketing and communications process. Once upon a time the Internet was this new thing that companies didn’t understand. PR outreach to online-only publishers and journalists was non-existent. Then came social media, and again we saw a long, painful process as confused businesses struggled to see how it fit into their marketing strategy. Mobile? Forget about it.
The same is true for search. At this point in time, SEO is still generally treated as a separate thing from marketing and communication. It’s related, sure, but is rarely, if ever, treated as a fully integrated piece of the marcom function.
This is going to change, and Google will make sure of it. Over time, it will be increasingly essential that search principles be embedded in the marketing organization from the point of conception. Instead of being an external function brought in after the fact to “optimize” the storytelling process, it will strategically inform the storytelling, product development and upstream marketing/research processes.
People who currently know nothing about search need to onboard a search-savvy view of their jobs. And on the other side of the equation, I think it’s going to be important for search people to begin conceiving of themselves as marketing professionals instead of as search specialists. This last bit is critical, because specialists don’t get invited into the strategic decision-making process very often. They’re regarded as tacticians only, and are rarely solicited for their opinions on anything other than technical concerns.
Search-infused marketing in a Hummingbird world must be strategic and it must have access to the C-level. If not, it’s going to have a hard time in the Google X.0 future.