The History of the GoogleBomb
And Google's Response To It, Part 2 of 3
June 20, 2011
Posted in Ken's Blog
Leave a Comment
We covered the first half of the 10-year history of the Googlebomb a few days ago. What do we have to show for it, so far?...
After 5 years, Google finally admitted the existence of a Googlebomb! There was, though, no ETA for a comprehensive resolution of the problem. Then came an event that many people remember as "the day Google fixed the bomb."
Their memory of this day is wrong. "Evolving opaque transparency" reaches new heights. As we shall see...
1) Google is the master of saying one thing (i.e., "we have begun minimizing," whatever that means), while...
2) Folks take away a different meaning (i.e., "Google kills bomb").
To this day, there is no ETA for a definitive Googlebomb fix. And Google no longer denies its existence. It merely denies that it's a problem.
Today marks the pivotal turning point in this 10-Year history. Let's pick it up 15 months after Marissa Mayer's "admit and trivialize" pronouncements (5.75 years ago)...
4.5 Years Ago
January 25, 2007
Google "Minimizes" The Bomb (by Ryan Moulton and Kendra Carattini in Google Webmaster Central Blog)
A mere 3+ years after "miserable failure" was reported, Google reports the bomb as being "minimized"...
"By improving our analysis of the link structure of the web, Google has begun minimizing the impact of many Googlebombs."
Note that it is not "fixed." Nor has it been "minimized."
Google has merely "begun minimizing the impact."
Google communicates at 2 levels. The superficial reader takes away the meaning that s/he is supposed to have. Most read at this level. Careful "Google-parsing," though, can reveal a very different story.
TIP FOR READING GOOGLESPEAK: Google gives the impression of open, casual communication. But no word or meaning of any significance is released without careful vetting.
With that in mind, let's review their announcement (above) in detail...
The word "prank" is used 3 times in this article (twice in the preceding article), including once as part of the definition of Googlebomb...
"... a prank where people attempt to cause someone else's site to rank for an obscure or meaningless query."
Serious journalists have better things to do than worry about the occasional childish joke played on the Web. With a wave of Obi-Wan Kanobi's hand, it mesmerizes journalists... "Move on, nothing here." ;-)
"Obscure?" Yes, in the sense that it is not a highly competitive query. But much damage is done (to searcher and victim) by commercial Googlebombs aimed at product trademarks, for example. No pranks there.
"Meaningless." Quite the opposite. According to Google, users of Google believe a misleading #1-ranked article. That can lead to making bad decisions. And to the victim of the bomb, that query is also highly meaningful.
Why does Google try to downplay the significance of the Googlebomb by using such language repeatedly? Obviously, to "minimize." It's not the bomb they are minimizing, it's the perception of its importance.
They are in fact not harmless.
Note the headline's wording (which is always carefully chosen).... "A quick word." They are implying that they are just passing along something quickly... nothing serious you understand?
It is important to learn how to parse "Googlespeak." This announcement actually marks the beginning of the "trivial-but-fixed" era.
This is a masterstroke. Not only do they continue to trivialize, this announcement will leave us with the impression that the bomb is finally fixed. Let's keep parsing...
2) "Under 100 Known Bombs"
"There are under a hundred well-known Googlebombs..."
If there were so few, why not make temporary hand changes until the "real algorithm" catches up? Why even bother with writing a special algorithm if you only have to "tag" a small number of bombs and be done with it?
i) There are many more bombs than the 100 known to the public (which can cause embarrassment for Google and doubt them).
ii) There are many, many more lesser-known bombs that are ignored (because they they do not cause embarrassment or doubt).
iii) The number of Googlebombs in the future will only increase. The technique still works and commercial bombs pay huge dividends.
And what's the deal about "improving our analysis of the link structure of the web?" Wasn't it pretty good already? This seems like an admission that the Googlebomb does beat Google's analysis of links.
I consider that to be important, not trivial.
As we shall soon see, the only bombs that Google considers to be worth fixing (i.e., eliminating, not "minimizing") are the "H-bombs," the ones that cause Google pain by hitting the prime-time news and front pages.
Those are fixed... fixed fast by a manual process, not algorithmically.
This announcement is merely a smokescreen that "minimizes" our perception of the importance and effectiveness of the Googlebomb, rather than "minimizing" the bomb itself. In other words...
We have entered the "trivial-but-fixed" era. Masterful...
While it's trivial, Google admits that it is actually worth dealing with since suddenly (?) Google seems to realize that people trust Google and assign come credibility to being found at it, something that the rest of us took for granted since the beginning of search engines...
3) "Google Validates Bombs"
"Over time, we've seen more people assume that [Googlebombs] are Google's opinion... it seemed like it was worth trying to correct that misperception. So a few of us who work here got together and came up with an algorithm that minimizes the impact of many Googlebombs."
This was preceded by an intriguing statement...
"Because these pranks are normally for phrases that are well off the beaten path, they haven't been a very high priority for us."
Is this saying what I think it's saying?...
Google does not care about the most flagrant type of violation of the vaunted Google guidelines (the very core of Google Search), as long as it's off the beaten path?
Google admits that these "pranks" are not so harmless after all. People believe "they are Google's opinion." And that gets Google's attention...
Google worries about how people perceive its search quality, which is the motivating factor to do something here.
In fact the gravitas of Google damages more than Google. It hurts Google's users who believe that the results are Google's opinion. It hurts the victims of Googlebombs (ex., competitors).
Trivial, serious - who knows what Google thinks? One thing for sure... Google thinks it serious enough to fix when it is the victim.
It would be nice to see a sense of responsibility by a company with a 2/3 market share of search. It would have been nice to see Google make a serious effort to eliminate the bomb because it violates the very core of Google Search and hurts others beside itself.
Google used to be different. It used to work to a higher moral bar. I'm afraid we won't find it defending principles again. Instead...
i) Googlebombs are both trivial ("pranks that are well off the beaten path") and getting serious ("people assume that they are Google's opinion").
ii) The Googlebomb is "minimized" (not "fixed"), and...
iii) It's all done as casually as if a "few of us got together"... over beers?
If this was as easy as "a few of us getting together," why not "minimize the impact" years ago? Does "hurt" to Google (i.e., users doubt Google's search quality) take priority over Google-users who make wrong choices due to this? Does it trump the "hurt" to hundreds, likely thousands, of other stakeholders (most of them less well-known)?
Apparently so. The interpretation here is...
"We've cured our pain. We have 'begun to minimize' yours."
4) "No Manual Fix?"
Why don't they fix these bombs by hand?
"You need to know a little bit about how Google works. When we're faced with a bad search result or a relevance problem, our first instinct is to look for an automatic way to solve the problem instead of trying to fix a particular search by hand. Algorithms are great because they scale well."
Six years of unfixed bombs (and various stages of Google cover-up) later, we're finally getting to the clear truth. Google does not "like" making manual fixes (as Ms. Mayer admitted 15 months earlier). So...
The heck with Google guidelines (the core of search), bad results "fed" to Google's users, and the pain that bombs cause to victims. Basically, they are saying...
We don't "like" to make hand changes because they don't scale. And that is more important than anything and everyone."
"Scale" is a wonderful word, so "techie" and important-sounding.
"Scale" is also a business term. It's code for "profit."
It's more profitable when any company "automates" high-volume events. Presumably, then, the Googlebomb is higher-volume than they are revealing. Or why bother? Just hand-fix bombs and "scale" for higher-volume problems.
Google likes to write algorithms, for whatever reason. The problem is (as we have seen since 2001 to 2007 and as we shall see right up to today)...
They either aren't very good at developing a scalable fix for the Googlebomb, or they don't try very hard. Judging by this Google announcement ("begun minimizing"), the latter would seem to be the case.
There was no claim of a "fix" here. They had merely "begun to minimize."
ETA for the Real Fix?
When are they actually planning to eliminate the bomb? There has been no real update in the 4.5 years since this January 2007 release, merely "H-bomb management" and the progressive evolution of Google's communications style of "opaque transparency."
As I noted earlier, detecting a Googlebomb is not hard (for Google) ..
Remember "The 5-Minute Googlebomb Algorithm?" It's really not too hard (for Google) to detect and defuse a bomb because the pattern of its development and existence is obvious. But 6 years after January 2001 (President Bush's 1st of 2 Googlebombs), Google cannot do better than...
"Let us know when it doesn't work"
"We wouldn't claim that this change handles every prank that someone has attempted. But if you are aware of other potential Googlebombs, we are happy to hear feedback in our Google Web Search Help Group."
They are disclaiming their non-fix before it has seen the light of day. Actually, there is no solid claim to have fixed anything. Whatever they are announcing, they expect it to be beat.
The only "minimization" appears to be the hand-picked removal of "well-known" bombs. Lists won't pick up future bombs, though. Only an algorithm will do that. And Google's disclaimers are letting us all know it.
And if you are the victim of a Googlebomb, you already know that the suggestion to "feedback to Google" gets no reaction at all, not even years later. Why?
Because there is no "Public Relations Return on Investment" in it for Google. If you happen to be suffering from a Googlebomb, get it onto CNN. Standard form letters will suddenly turn into immediate action.
Bottom line? Most of us remember this January/2007 announcement as "the fixing of the bomb." Nope! It all boils down to nothing more than hand-fixing a list of well-known Googlebombs for public relations reasons.
Google remains beatable by new bombs, while users have bad experiences and while victims of lesser-known bombs remained victimized. And that's OK with Google...
"Forget the Google guidelines. Let Google users suffer sub-par experiences, including the bad recommendations of commercial bombs. Let bombers win while victims suffer."
This announcement, it turns out, has actually "begun" very little.
The Special Danger of the Commercial Googlebomb
According to this release, Google finally realizes that searchers consider these bombs to be Google's opinions. They must understand the implications, too...
1) Surfer searches for product "XYZ."
2) While entering "XYZ," Google Instant (or Recommend) offers "XYZ scam" or "XYZ sucks" as the top option.
3) These are eminently bomb-able terms (i.e., they are relatively less competitive). According to Google...
4) A searcher at Google assumes that a fake negative review or article is "Google's opinion!"
5) Merchant loses sale. Searcher buys wrong product.
Commercial Googlebombs are particularly dangerous...
- The profit motive is more powerful than the "prank" motive," encouraging more and greater efforts.
- These bombs are allowed to persist, since they rarely make the prime-time news, with exceptions such as when the New York Time broke the JC Penney story (you can see my discussion of that here).
- They work. A surfer trusts Google to deliver the best results possible. S/he believes a nasty "review" to be Google's opinion, not the result of a planned manipulation by unscrupulous competitors
- The damage is particularly high because searchers who search for a product or company name are close to making a purchase decision. Conversion Rates of victims plummet, causing financial damage to the company.
Googlebombs against company or product trademarks will become increasingly common as the cost/benefit ratio is understood. Consumers increasingly search these terms when seriously considering purchases. They search for "reviews" or "scams" related to such terms.
If you were a competitor with a rather low moral barrier to entry, why not get in front of that highly motivated audience? What do you think happens when they read the #1-ranked review that is a Googlebombed result of your fraudulent efforts?
It carries "Google credibility," as admitted by Google.
The Google user reads paragraph after paragraph of negativity and then finds more links to other so-called "reviews," who are all "in on the bomb."
The natural human reaction is, "Where there is smoke, there is fire," especially if Google seems to be fanning the flames.
That's no "prank!"
Is it right that Google fixes 100% of the mass-media-publicized Googlebombs, the ones where Google's own image suffers?
Is it right that Google encourages its users to make wrong choices and that it leaves the "lesser-known" victims/products to flounder?
Why Doesn't the Algorithm
Protect You Against Being Bombed?
Lesser-known bombs may not be front-page "USA Today," but they are still often blindingly obvious. If Google's "fix" was indeed algorithmic (as they claim), these simple bombs should be detected and defused (i.e., de-ranked). They should never see the light of day.
It seems, however, that the "well-known" are the ones that were fixed. Their algorithm fails for the lesser-known, unknown and new ones, so basically it "minimizes the effect of a hand-created list of well-known bombs."
And that type of "algorithm" is not really an "algorithm" (in the sense of my 5-minute algo that can identify and neutralize all types of Googlebombs). It is basically a list of manual changes. Let's continue on this line of analysis...
If the Googlebomb really had been fixed "algorithmically" (ex., see my "5-Minute Googlebomb Algorithm"), other bombs should have been eliminated. A true algorithm is effective in finding unknown bombs and preventing new ones...
And it would not be hard for Google to execute.
But when you look at their 2001-2007 history, "culminating" in this release by some guys who "got together" to "minimize" the bomb, one has to wonder just how committed Google is to fixing this. Could their algorithm consists of nothing more than a list of those well-known bombs to eliminate?
One has to wonder because if an algorithm existed that 1) detected and 2) eliminated, shouldn't obvious ones have been detected and stopped? Shouldn't new ones stop appearing? One would think so...
More bombs detonated after that, including some great big noisy explosions. This one, where "dangerous cult" sent folks to Scientology.com's website, was reported 1 year later, on January 29, 2008!
Needless to say, the church is strong enough to create enough noise to get Google to "minimize" that one pretty quickly. Wonder where the algorithm was, though?
Could the "Scientology bomb" have succeeded if any sort of effective algorithm actually existed?
Come to think of it, if Google could claim that they have an algorithm that they run "infrequently," that would make a pretty darn good excuse.
Wait, I'm getting ahead of myself. Google had not yet reached this stage of "evolving opaque transparency." ;-)
SEM (Complex) vs. GBO (Simple)
Consider the huge difference in the history of SEO ("Search Engine Optimization") vs. the history of GBO ("Googlebomb Optimization"). OK, OK, I made up the term "GBO!" But that's the point...
The history of SEO is one of increasingly sophisticated manipulations to rank high in response to increasingly sophisticated search algorithms (for "black hats," not "white hats," but let's not split hairs over this). It's an arms race between SEOers and search engines. It is becoming increasingly difficult to manipulate search engine rankings through SEO.
(For more on the limited future of SEO, right-click here to download "The Tao of C T P M".)
On the other hand, successful "googlebombing" is virtually no more sophisticated than it was 10 years ago. My "5-minute spec" would detect-and-defuse as well today as 10 years ago.
And that means the following...
If an algorithm really did exist to meet those simple specs, Googlebombs would be stopped (or would require such a level of effort that they are not worth the effort). So then...
Why do the same old Googlebomb techniques still work?
Simple. There is no algorithm. Sure, Google can say one exists. Depending on how one defines "algorithm," they might even consider themselves truthful.
However, there is no algorithm, not one that is equivalent to "The 5-Minute Googlebomb Algorithm." The facts are simply not consistent with its existence.
This runs contrary to a lengthy article that was published on the same day as Google's blog announcement...
"Kills" is not the same thing as "begun minimizing."
SearchEngineLand (SEL)'s article opened categorically...
"After just over two years, Google has finally defused the “Google Bomb” that has returned US President George W. Bush at the top of its results in a search on miserable failure. The move wasn’t a post-State Of The Union Address gift for Bush. Instead, it’s part of an overall algorithm change designed to stop such mass link pranks from working."
This goes well beyond what Google claimed. SEL has rapid access to senior people like Matt Cutts for quotes such as...
"It's completely algorithmic."
Cutts even gets the escape hatch into the SEL coverage...
"We're not going to claim it's 100 percent perfect."
Nothing is 100%, of course. It's unusual, though, for Google to announce something new with "nothing is 100%." And, after all, a well-constructed algorithm should eliminate most Googlebombs. But an algorithm that covers a list of known bombs won't.
It sounds like Google is anticipating new bombs showing up. And for good reason, as we'll see in the third and final installment.
So Where Do We Stand?
It seems that the Googlebomb was not so "minimized" after all. I doubt how much they had actually "begun."
The structure of a bomb is so gross that any half-decent algo would disarm them all, with ease. Google claims to have a fix. It seems to be a manual fix, that only seems to work for the ones that make the media (and other "well-known" ones).
The lesser-known were not "minimized," and new ones kept springing up.
Worse, detailed feedback to Google (as requested by Google) by victims of Googlebombs get only form letters in return. No action.
TIP: If you are the victim of the Googlebomb, get it on TV! ;-)
Stay tuned for the third and final installment. Matters are heating up...
Fireworks (and Googlebombs) will light up the skies and Google will take "evolving opaque transparency" to new lows (or highs, I suppose, if you're Google ;-) ).
Really though, why not just do the right thing?
All the best,