Do you have to “be someone” in order to have an impact on search positions in Google default search? Maybe not today, but its certainly looking like that’s where we’re heading. It used to be that you could just do some information architecture and hyper-optimize a site as some anonymous guy in tiny server closet and have a big impact on sales. But today, in the fight against spam, evidence suggests actual people are going to have to put their reputations on the line to vouch for content.
Regardless of how search evolves, chances are that “generic search” will likely continue to exist, because we don’t always know or take the time to choose a specialized search starting-point such as Yelp or Amazon. We just want to start typing or talking and have the best results come back. And so long as this generic type of search continues to exist and be a popular starting point, there will have to be some sort of relevancy and ranking system to sort it all out.
You might think that explicit data sources and API’s that can insert Yelp results for restaurants or Amazon results for shopping could change all this generic search ranking. But no, it’s actually the opposite. The mountains of Web data becomes unimaginable volumes of structured pure data. And guess what? Even that API stuff needs to be sorted and ranked. No matter how you slice it, you can’t get away from relevancy ranking systems. Therefore, what we think of as the field of search engine optimization (SEO), while it may change dramatically, its not going away.
The problem is that the old fashioned way of ranking and sorting things on the web is too easily corruptible, because the things themselves can be created, destroyed, copied, algorithmically tweaked, etc way too easily. Such “things” include registered domains and the websites that are hosted on them. You might say that is not the case with Yelp and Amazon reviews, and you would be right. That is because the reviews hosted there are already tied to hard-won reputations that people are putting at risk every time they write a good or bad review. This approach is simply being extended to more things.
Because there are a finite number of PEOPLE as opposed to THINGS, it makes us an ideal starting point for sorting out the relevancy of things that relate to us. When I was a kid I remember the world population number going from 4 billion to 5 billion. Today, it’s 7 billion. Now that seems impressive, but the fact is that there is that people are a scarce resource, and new ones are only being issued at a relatively slow rate—especially when compared to Moore’s Law. Today, a cheap hard drive typically has a trillion bytes—providing approximately 100 bytes per person. Most characters fit into 2-bytes, making 50 characters per person on a $100 drive. In other words, a cheap hard drive can contain the email address of every person on the planet (if everybody had one).
That’s just one hard drive. A bank of hard drives has enough storage to keep a social affinity map of the relationship between people and things for the entire planet—a list of URLs liked by each person. Forget about data centers. Once you adopt people as the “master node” in a link graph and normalize your URLs, the data behind global-scale every-person-on-the-planet tracking systems could fit in one rack. Sure, a real application is more complex because of bandwidth and concurrency—but speaking purely about the data behind it—it’s not so big, if you start with people and normalize everything they connect to.
And on these social graphs, if you want to be an influential marketer, you’re either going to need to be an extremely well connected node on that people-link graph, or you’re going to need to know how to influence people who are those hubs. This is taking the place of the old domain-and-page link graph, because as opposed to people, new domains and pages can be issued at an infinite rate, far surpassing the ability to accurately track and fight manipulation, even factoring in Moore’s Law. The information explosion is happening at far faster rate than Moore’s Law. The opinions of PEOPLE therefore becomes the gold standard of measuring value of things.
So people like me who were used to—and quite happy with—being relative unknowns and still having an impact have some decisions to make. We can either remain unknowns, and specialize in the what will remain of old-school SEO by which we fine-tune long-tail landing pages, trying to make them into the best possible page on that niche topic on the planet. This is a valid option, because just like I believed in 2006 when I made HitTail for exactly this purpose, I still believe there is unlimited small bits of business that can be collected together into something worthwhile. It’s just not quite the gold-rush it used to be—and perhaps not of as much interest to big advertisers where they can just purchase 100x more traffic with 1/10th the effort.
Or you can become an influential hub in the new social link graphs. There is a built-in problem with this. It is likely that you will only achieve influence over search rankings in one or two fields on which you have been identified as an expert. So, if you are a marketer, you will become known for marketing—and if you are an SEO, you will be nice and findable by people looking for SEO’s. However, in your capacity as a marketer or SEO, you will be unlikely to be able to directly directly elevate your client’s rankings on their keywords by +1′ing their content.
So, if your client is a chain of hotels, you would be in the awkward position of… perhaps WANTING to +1 or review or rate each hotel in the chain. However, this is obviously not a workable approach, so your role as a marketer is to ensure that OTHER genuine seasoned, well-known reputable travellers who are staying at these hotel chains go ahead and review, rate and +1 the content for the hotel they’re staying at. It’s even more complicated now, because that content probably resides both on Goolge+ Local AND the client’s own site… and of course also Tripadvisor. Now that Frommer’s Travel Guides has been bought by Google, it’s scoring data and system will probably be wrapped directly into Google the way Zagat’s was.
Hotels is an interesting example, because it shows that relevancy signals and scoring mechanisms differs based on the type of thing being searched. There is also questions around where the data lives and who owns it. In the case of Zagat and Frommer’s, Google bought the data and the scoring system and the relevancy factor, and is wrapping it natively into Google’s systems. But the HTML5 microdata and schema.org standards (and older formats) actually allow similar data to reside self-published right on the site-owner’s own site, such as those rich snippet 5-star aggregate reviews. Of course, since it’s publisher-provided, it’s ripe for spamming.
But this site-published metadata can be less spammy and more trusted if “signed”—either by the verified publisher of the site, or by each individual person doing an individual review. Aggregate reviews are “rolled up” individual reviews, so each individual review would have to remain accessible on the site, and Google or whatever crawler, would have to go in and double-check the validity of each review. But then, for symmetrical verification, each person making a review, would also have to put THAT SITE in their Google+ profile page under “Contributes to”. You would put such sites as Amazon, Tripadviser, Yelp, or the manufacturer’s own site if reviews are a feature of their own site. This could get messy, which is why I guess Google bought Zagat and Frommer’s.
Inevitably, identity keeps creeping into the picture. And when you’re a company with an online brand, you can’t very well post as yourself, or call yourself the publisher on the brand’s site. The company itself sort of has to exist as a person online. And so in its patent addressing how people will influence search rankings, Google called it AgentRank. They resisted using AuthorRank (instead of AuthorRank). The word “author” is too indicative of a particular profession in a particular industry: writers in publishing. And while sure, a lot of the Web is structured around the publisher model with identified authors, much is not. Much is just the company’s website who don’t have writers as-such. And so, “agents” can be working on behalf of a company. Profiles are created FOR the company, but are very similar to user profiles, and can be verified as genuine identities.
And so, there is yet another role that the next-generation SEO can play: helping companies navigate all that complexity when identities come into the picture. Identities tie-to data, validating it, and therefore allowing it to be used with a bit of confidence by Google as a ranking factor. So, there’s some technical site preparation work to be done: good URLs, good metadata, proper publisher verification, and good symmetrical signing.
In other words, if you can’t use your own reputation to directly help boost your client’s rankings in the hotel space because you’re known as an SEO and not a prolific traveller, then you can at least help clients set up their sites properly, and sheppard their loyal customer flock along the star and review-writing path. SEO’s tend to revel in their barely understandable technical complexities and difficult-to-pinpoint cause-and-effects, so this is an area of specialty that I suspect will appeal to many from the old school of on-page/off-page factors.
There is a line of thought that the Web world is running headlong into the API world, and that the Web the way it was during the 90′s and 2000′s that relied upon a crawl to sort it out is sort of old-fashioned. Any data that’s worth anything—reasoning goes—will be available in an explicit data feed. This is the proposition of Wolfram Alpha, and it is what “assistant” systems like Siri use to make the first set of answers best, before “resorting” to a Web search as the less desirable backfill. Well, doing SEO for this API-centric world is very similar to helping clients navigate the complexity of an identity-controlled world. Putting data in an API doesn’t solve the fact that the data will be too voluminous to make sense of without some sort of scoring and ranking system that somehow ties back to identity. And so, we get back to identities.
And that leads to a third—and to me, most interesting possibility. The first two possibilities being relegated to the long tail or being an API / identity technician. Anyone worth their salt is going to be ascending to become a domain expert in something. Paraphrasing Seth Godin: if you’re not trying to be the best in the world at something, why are you even trying? And so, focus in on your passion within SEO. Become the best in the world AT JUST THAT. Publish prolifically. Find your audience online. Stay aware of what others in your field are doing, and +1 and comment on THEIR content when it merits. Make sure they know who YOU are, and start connecting the dots that need to be connected.
Do that thing that you do better than anyone else, and just make sure you’re optimizing your social behavior and content around that—so, you sort of have to be a little bit of an expert at approach #1 and #2 anyway. Keep fishing for opportunities on the planet at large—your market. The world is a very small place now, and if you are great at what you do, and there is any market for it, you and that market WILL BE CONNECTED.
Problem: people are not single-dimensional beings, and it’s hard to choose what you WANT to be known for. It’s not like you can control the keywords someone used to name the circle they dropped you into—unless you’re Danny, Matt or Rand, in which case it’s obvious. But at least you can be dropped into multiple circles by one person, so you have some leeway to work with—and some of your keyword authority will probably come from your signed content. But still, you’ll probably only be able to ascend to expert status on a couple of things—just as in real life?
Seth Godin put out one of his brilliant little blog posts recently about resisting being categorized. He was speaking in relation to your business offerings, but because personal reputations are becoming so important online, he might as well have been speaking about personal brands. He says that you should pick a standard category for yourself, because people have too much going on to deal with new categories. Human beings categorize, because that’s how we view the world to deal with information overload. If you try to do something new, people are going to categorize you anyway, and you’re just going to lose whatever control you may have had over it.
And so that leads to the subject of what will inevitably be my next blog post. How do you choose your own categories when making your personal brand online? Is that the same thing as your main navigational menu items? What should those items be? Should they be as broad and multifaceted as you are as an individual? Or should you pair it down to what you know the public at large is most interested in you for? This and more I will address next.