Google Transition Rank: We’re All Spammers Now (part 3)

In the Search Engine Academy blog, we left off in the last part of this series with discussing some implications for actual implementation of the Google “Ranking documents” patent.  The third consideration to discuss is as follows:

3) Some serious thought will also need to be given to how this would be implemented.  Will just any on-site or off-site change at all be enough to trigger transition rank, or only certain ones?  Will any site at all be vulnerable to it, or only certain ones?  Where is the balance or threshold for Google in combating spammers without hurting its own SERPs?

This really gets at understanding to what extent we’d actually have to worry about transition rank if the patent is indeed implemented.  If any on-site or off-site change at all would be enough to trigger transition rank, you would have search results bordering on chaotic from the perspective of the searcher.  This is not good for Google and, in fact, counterproductive.  The purpose is to catch spammers; so what triggers transition rank would most likely be limited to the kinds of changes that look unnatural or spammy and to the kinds of sites that fall below a sufficient level of trust or authority to make it less suspect.

There are some spam tactics that are obvious like keyword stuffing, invisible text, skewed backlink profiles and the like.  Some really require intent and others could be done accidentally.  Because some could be done unintentionally, you really need to have some understanding of what constitutes spam and why so that you stay out of trouble.  It’s not enough simply to know what some good practices are.

Such is the Power of Wine

If, as an SEO consultant or as a webmaster, you view the role of SEO as tending to the competitive health of your website, you should take a helpful reminder from Hippocrates (the ancient Greek physician): “ἀλλὰ τίνα τε πόνον καὶ διὰ τί καὶ τίνι τῶν ἐν τῷ ἀνθρώπῳ ἐνεόντων ἀνεπιτήδειον”

Hippocrates

Hippocrates c. 460 BC – c. 370 BC (father of western medicine)

“It is not sufficient to learn simply that cheese is a bad food, as it gives a pain to one who eats a surfeit of it ;  we must know what the pain is, the reasons for it, and which constituent of man is harmfully affected. . . . Thus, to illustrate my meaning by an example, undiluted wine drunk in large quantity renders a man feeble; and everybody seeing this knows that such is the power of wine, and the cause thereof; and we know, moreover, on what parts of a man’s body it principally exerts its action; and I wish the same certainty to appear in other cases.” -
Hippocrates, De prisca medicina 

Applied to our situation, you need to understand not just what spam is, but also why it is spammy and what search engine principles (signals, algorithms, etc.) are affected.  Some tactics are just downright bad: invisible text, keyword stuffing, and the like.  Others can be a matter of doing the right thing in the wrong way.   In some of these cases, what gets designated as spam and might therefore trigger transition rank can be a matter of degree or proportion – like link text.   By simply maintaining a more natural distribution of link text you can neutralize the spam effect.  In other cases, you just need to remove the offense as with keyword stuffing and that will eliminate the spam effect.  In most cases, real trust and authority around your site (inherited or direct) will be your best friend in avoiding a problem with transition rank.  Spam will make you more vulnerable to transition rank.  Whether you neutralize it or eliminate it when found, or take preventative measures against its effects, dealing with it will make you less susceptible to transition rank.  Such is the power of spam.

If the “Ranking documents” patent is not implemented, all of the recommendations and tips from this series are still worthy of your consideration because they will still help your SEO strategy.

Learn about in-person, hands-on SEO workshop with one of our expert trainers!

Google Transition Rank: We’re All Spammers Now (part 2)

We left off in the last part of this series with outlining some implications for actual implementation of the Google “Ranking documents” patent.

There are some serious implications to consider however.

  1. There are difficulties for the relationship between SEO consultants and their clients.
  2. There are also some difficulties for doing the work of SEO itself, namely, measuring the effectiveness of your strategy.  In both these areas, the difficulties can be overcome.
  3. Some serious thought will also need to be given to how this would be implemented.  Will just any on-site or off-site change at all be enough to trigger transition rank, or only certain ones?  Will any site at all be vulnerable to it, or only certain ones?  Where is the balance or threshold for Google in combating spammers without hurting its own SERPs?

These are at least three areas of implication worth discussing and they are, of course, interrelated.

Lets go through these.

SEO consultants and their clients

For the SEO consultant, things get a bit complicated because a client may experience a drop in ranking/traffic after taking advice that was intended to improve things.  The consultant is in the position of having to convince the client that the decline is not only temporary but also that its duration will be uncertain, but that ultimately things are going to improve.  The client is in the position of possibly not being able to tell the difference between good advice from which the benefits will eventually be realized and bad advice from which that decline is the end of the story.  A consultant will have to be much more diligent and transparent in setting client expectations so that the need for patience and the areas of uncertainty are clear.

Building trust and confidence in the partnership will be even more important.  The consultant and client would do best to identify “low impact” areas of a site where SEO strategies can be  implemented first.  You can implement the strategy.  Then when and if transition rank is triggered, the financial impact for the client is minimal.  Once the benefit is ultimately realized, the client will have confidence in the strategy and therefore more patience with that strategy when applied to areas of the site where transition rank has a greater impact financially.  You can also build this kind of trust and confidence by showing a track record for your strategy with other clients whose sites and competitive landscapes are similar enough in the view of the new client to establish the comfort level needed.

Measuring the effectiveness of your strategy

Transition rank would clearly make measuring SEO effectiveness more challenging.  In part one, we raised the following important question.  ”How would you distinguish the characteristics of this patent in the behavior of the SERPs from other influences that might look similar or even very similar?”  You perform your website analysis (on and off-page factors), you devise your strategy, you execute your plan and then watch for the results.  But you know that many things influence the results including but not limited to:

  • localization
  • personalization
  • competitive landscape fluctuation

So if you see a decline in the short term, how do you know whether it is transition rank to which you should not be reactionary as opposed to a fluctuation in the competitive landscape to which you should respond so that your competitor does not get the upper hand?  Let’s presume you’ve got no alerts in webmaster tools or anything of that sort.  You need to understand enough about your competitive landscape to know whether the extent of the decline is warranted by changes in that landscape.  If not, then it is more likely that something else is at work.  The principle is fairly simple, but not so easily done in practice.  There is much that may be unclear or uncertain.  You could call it SEO quantum uncertainty.  Some parallels from quantum mechanics are actually helpful here.

SEO Quantum UncertaintyObserver Effect

This is a problem for optimizers.  With the phenomenon of Personalized Search, there is no set ranking or ordering of the competing pages by Google until a particular searcher sees the results.  In a sense, the ever fluctuating Google wave collapses only once a SERP is returned to a user.  The solution requires on-page and off-page semantic optimization.  Why and how is beyond the scope of this post but you can ask your local search engine academy instructor.

Quantum Uncertainty for Optimization

Beyond the observer effect, the optimizer encounters uncertainty when examining either an SEO factor or its weighting to ascertain its influence on the search results.  Due to the facts that 1) everything in the competitive landscape is graded on a curve and 2) that the landscape is constantly in flux, the influence of the factors and their weightings are also constantly in flux.  The solution requires better capturing the relationships between the factors and their weightings by using multivariate analysis.  This is not a new idea in SEO.  It has been well known and discussed as long ago as a 2007 Search Engine Guide article by Claudia Bruemmer.  You just need to put it into practice.

Optimizer Effect

This is a problem for Google.  As Google reveals to the public, or as optimizers discover on their own, what signals are being measured in its algorithms, the behavior of optimizers changes in a way that tends to result in a decrease in the effectiveness of using certain kinds of signals as a measurement by Google.  As a result, Google’s system must change and adapt to maintain the purity of its results.  Revealing some signals to the public is in some cases helpful for Google.  Having the Internet ecosystem take on certain characteristics (more people solving cannonicalization issues, more webmasters using Title tags properly, more webmasters using Authorship tags, etc.) makes it easier for Google to perform its analyses.  Some of those characteristics may not sufficiently materialize, however, if webmasters don’t know they are important.  Many of those desired characteristics are neither intuitive nor natural for webmasters like rules for cannonicalization, use of meta tags, and various SEO “best practices.”

Keeping these things in mind and applying them would help you distinguish between transition rank and other influences on the SERPs that could look very similar so that you have better confidence on knowing when to act and when to be patient.  Remember, none of this is some attempt to uncover a search engine’s algorithm.  It’s all about precision in measuring from competitive landscapes those factors over which you can have some influence.  The next part in this series begins with discussing how the “Ranking documents” patent might be implemented.


Addendum for Geeks

Quantum Mechanics - SEO Analogy

Formula - SEO & Quantum Physics

Some find it helpful to have a formula or visual aid to get a better handle on things.  To that end, the following figures are provided to help with understanding the analogous relationship between SEO and quantum physics.

Looking at a formula here related to Heisenberg’s uncertainty principle,

1) position,
2) momentum,
3) and angular momentum

would correspond respectively to

1) ranking position,
2) fluctuations in competitive landscape and algorithms,
3) and competitive momentum

Competitive momentum (CM) takes into account your average or probabilistic ranking along with your strength within a competitive landscape and gives you an idea of how stable your position is as well as the direction in which you’re moving for a particular URL and a specific keyphrase.

If you are really curious, below is how the formula might look to calculate competitive momentum for a specific URL over a set of K keyphrases:

Competitive Momentum for a group of keyphrases

Competitive Momentum for a set of K keyphrases

Learn about in-person, hands-on SEO workshop with one of our expert trainers!

Google Transition Rank: We’re All Spammers Now (part 1)

Transition Rank

Google "Ranking documents" Patent

Google was granted a patent in August of 2012 that may have a significant effect on a critical component of SEO research and tracking. It is officially called the “Ranking documents” patent (read a great summary of the Ranking document patent by my friend Bill Slawski). Some in the SEO community have been referring to it as “Transition Rank” and also the “Rank Modifying Spammers Patent.” Many think they have seen evidence of it already at work. After on-site or off-site changes are made for a site, this system toys around with someone’s ranking to see if there is reaction by attempting to make corrective changes. If so, the site may be designated as spam. In an illustrative scenario in the patent, they suggest such a document could lose ranking for 20 days before beginning a 70-day climb to the NEW-Rank. So one could make a good change that should result in increased ranking, but this system would demote that ranking first before gradually putting a page in its rightful, new, better ranking position. I plan to discuss several issues surrounding this patent including:

  • Reactions to this patent in the SEO community
  • What its implications would be if implemented
  • Some insight from a different angle on this relatively old news
Reactions in the SEO Community
“This is a messed up situation because even if you change what you were doing to follow the line of Google’s acceptable SEO practices, it is still viewed as an attempt to modify Google’s index and thus it is the work of a spammer.” – Tom Forenski

Even "white hat" is "black hat"

This quote from Tom Forenski nicely sums up the reaction from many in the SEO community to this patent.  This would make everyone engaging in SEO a Spammer!  ”We’re all spammers now!” would be the collective mantra of webmasters the whole world over.  As you walk down the street, “Oh, you’re one of those dirty ranking modifiers aren’t you?!” is the accusation you might hear just for cleaning up your title tag.  In your next webmasters meetup group, don’t even let it be known that you engage in off-page optimization.  You’ll find yourself being unfriended, unliked, and removed from circles by people just to avoid digital association with you and avoid linking to a bad neighborhood, regardless of how ironic such a response would be.  Is the situation in which we now find ourselves really this grim?

It’s a red herring!

Is this just propaganda on the part of Google to influence the actions of anyone engaging in SEO?
The thought here is that this patent is not anything that Google actually intends to incorporate into its combination of algorithms and signals for ranking web pages.  It’s role is simply to send SEOs running like chicken little.  You might call it a disinformation campaign or counter intelligence against aggressive SEO practitioners.  This would constitute a counter measure on the battle field of adversarial information retrieval.  Keep those spammers so afraid that you’ll send them running.  Only in this scenario, almost anyone doing any optimization at all might be categorized as a spammer (at least under some understandings of the patent’s anticipated implementation).  All the respectable webmasters will begin to warn you, “Optimization is the path to the dark side. Optimizing leads to rank modifying. Rank modifying leads to transition rank. Transition rank leads to suffering.”  If you buy all of that, you’ll definitely stop doing SEO.  So as a scare tactic, it could be very effective on the part of Google.  However, is it wise to reckon it as just a scare tactic devoid of reality out of hand?  Okay, maybe you’ll do some research on it first; conduct a test; analyze some results.  But how would that look?  How would you distinguish the characteristics of this patent in the behavior of the SERPs from other influences that might look similar or even very similar?  I’ll have something to say on that question a little later.

They’re out to get me!

Is this some sort of conspiracy to destroy any and all SEO activity through death by confusion?  Need a tin foil hat, anyone?  ”Hey!,  just because I’m paranoid doesn’t mean no one is out to get me.” might be a clever rejoinder here, and yet strangely true.  Is this Google’s attempt to shut down the SEO industry and drive almost everyone running to PPC as the only reliable means of generating traffic from the SERPs? Is, dare I say it, SEO really dead this time?  Is this time “The king is dead” not followed by “long live the king?”  Rankings would become so impossible to interpret (by those who still obsess over such things) that many might abandon the task of optimization altogether.  Is Google really that bent on confusing and sticking it to optimizers?  Well, if you’re an algorithm chaser, you may already be your own worst enemy in this field even apart from the “Ranking documents” patent – implemented or otherwise.  Even so, Google is out to get spammers.  They have to be for all the obvious reasons.  However, for the same reasons they have to fight spammers, they can’t monkey around with their SERPs too much.  They need fast, reliable, and relevant results so surfers keep coming back to use their search engine and click on those ads.  Google needs traffic as much as you do.  They can’t afford to cut off their nose to spite their face.  Even if they are out to get you, they can only go so far in doing so.  This, too, I will address a little later.

The implications if implemented.

What would it really mean for SEO if the “Ranking documents” patent is incorporated into Google’s algorithms and signals for ranking web pages?  From now on, when anyone considers entering the world of SEO, will they see the following inscription on its gates?
Abandon all hope, ye who enter here.

Lasciate ogne speranza, voi ch'intrate

“Lasciate ogne speranza, voi ch’intrate”

- “Abandon all hope, ye who enter here!” (Inferno, Canto III, line 9)

I don’t believe that properly characterizes our situation; not a bit of it.  There are some serious implication to consider however.  1) There are difficulties for the relationship between SEO consultants and their clients.  2) There are also some difficulties for doing the work of SEO itself, namely, measuring the effectiveness of your strategy.  In both these areas, the difficulties can be overcome.  3) Some serious thought will also need to be given to how this would be implemented.  Will just any on-site or off-site change at all be enough to trigger transition rank, or only certain ones?  Will any site at all be vulnerable to it, or only certain ones?  Where is the balance or threshold for Google in combating spammers without hurting its own SERPs?  These are at least three areas of implication worth discussing and they are, of course, interrelated.  Part 2 of this series picks up there.

Learn about in-person, hands-on SEO workshop with one of our expert trainers!