Search Optimization and Its Dirty Little Secrets | Hacker News Reader

Search Optimization and Its Dirty Little Secrets

nytimes.com

91 pointsatularora30 comments

There's no dirty little SEO secrets!

Tagged with seo, dirty, dirtyseo, secrets, littlesecrets, little secrets, dirtylittlesecrets, dirtyseosecrets, dirtyseolittlesecrets, dirtylittle secrets, dirty littlesecrets, dirty little secrets, dirty seo, dirtyseo, seo secrets, seosecrets, secrets of seo, dirty secrets of seo.

PS. Don't forget to link companies, products and names to your own sites, and remember if you get caught putting two external links in an article you get fired!

<3 Engadget, and most pro-i-cant-believe-this-is-what-was-supposed-to-replace-journalism blogging.

"I think we need to make a distinction between two different kinds of searches - informational and commercial," he said. "If you search ‘cancer,’ that’s an informational search and on those, Google is amazing. But in commercial searches, Google’s results are really polluted. My own personal experience says that the guy with the biggest S.E.O. budget always ranks the highest."

Perhaps, for predominantly commercial searches, the old 'goto.com' (which became Overture) had it right: top results should be blatantly (and honestly) paid-for.

Google could implement this by showing more paid results for commercially-motivated searches – while allowing users to dial this back down if desired. (They do this a little with the varying number of AdWords units above natural results.)

Pretending the top results are not going to the largest budget, simply because they're the result of massive investment in gray SEO rather than direct payments to Google, is in some ways the worst approach. It enriches some shady operators, and has negative externalities polluting the rest of the web.

At one point years ago someone – I think it was Yahoo – experimented with a slider atop results that let you bias your results in a more- or less-commercially-oriented direction. It might be time to try similar experiments again, with a new UI and the now-more-sophisticated searchers of today.

Google has every economic incentive to prevent paid SEO strategies from working. Google sells placement on the results page--that is their whole business. Gray- and Black-hat SEO is the only serious competitor; money Google takes away from those operators will go straight into Adwords.

At a surface, conscious level, you're right. Google wants any tolls paid to appear in their results to be paid to them.

But considering the whole competitive ecosystem, and indirect effects over time, nasty SEO can help make Google money. Consider these mechanisms:

• when the natural results get polluted, the vetted AdWords placements become relatively more attractive to clickers and advertisers.

• Google' self-image requires they reserve some ostensibly 'non-paid' slots on the front-page. If in practice these all still have a high cash cost, even though paid to third parties, then everyone needs to open their wallets, rather than just do pure-of-heart helpful info-design, to appear on the front-page. That is, amounts paid to third parties serve as a short of shadow bid, placing a rising floor on the value of AdWords placements.

• SEO tactics that mildly or temporarily fool Google, but that Google then overcomes, provide a dynamic, ever-growing barrier-to-entry to other later-entrant search engines. Competitors don't have to do as well on the same web Google first mastered; they have to do as well on the web mutated by Google's economic incentives and iterated cat-and-mouse game.

> Perhaps, for predominantly commercial searches, the old 'goto.com' (which became Overture) had it right: top results should be blatantly (and honestly) paid-for.

But that is more or less the case now - the top commercial results are paid for (you need to pay for SEO). If you will pay Google directly for that, who would use Adwords then?

Google does this already - compare [flowers] with [ipad] with [ghc syntax]:

http://www.google.com/search?q=flowers

http://www.google.com/search?q=ipad

http://www.google.com/search?q=ghc+syntax

There used to be "More/fewer shopping results" in the left nav as well, but these seem to have been unlaunched.

Thanks for the examples; I did mention the varying number of top AdWords units as a small step in this direction. I'm suggesting the best defense against SEO pollution might be to take this even further.

That's not Google's doing, that's just representative of the adwords inventory. Obviously, there is no advertiser anywhere who's going to pay for the keyword "ghc syntax".

I work with Doug Pierce, who is mentioned in the article. If any HNers have any questions about this story, I'm happy to answer the ones I can and pass on any that I can't.

How did J.C. Penny's get singled out for the story?

Size and egregiousness, I think. There are worse offenders, and there are bigger companies that play fast and loose with SEO, but this is the biggest company to do something so black-hat.

How was it detected? Just noticing Penney seemed to be ranking awfully high in a lot of searches, or something else? (I ask because the blackhattery mentioned in the article didn't seem very noticeable in its own right.)

Yeah, just noticing it.

I actually figured out what the basic strategy was, but Doug dug up a ton of data on exactly what they were buying. Some of it is kind of insidery stuff you'd want to know if you were an SEO but not if you were a typical Times reader. (David Segal does a really impressive job in general of writing about this stuff in a way that's understandable to the average person. I hope he does an article about white hat SEO some time, so I can cite it when I explain SEO to my clients.)

It may be more black-hat than some attempts, but the most frustrating thing about this article is that it implies that paid links are a rare thing for e-commerce sites. Pick any shopping-related term, and the odds are overwhelming that the top-ranking site has a large SEO budget.

Agreed, if the average-user only knew how prevalent this is. We're an e-commerce site and constantly feel in the minority for not buying links exactly like JCP did. Almost all of our top competitors do the exact same thing and have gotten away with it for as long as I can remember. The only reason we don't is being too afraid of the supposed "wrath" that doesn't ever seem to come to them (that is unless you can get a NYT article written about it).

Interesting that the New York Times seems to have done all of the footwork in finding & reporting JCPenney in order to have a ringside seat to the public hanging. There is only really a passing mention of the NYT's direct involvement... I don't have any problem with them outing an obvious example of manipulation with paid links, but I wonder if they are unsure how some will react.

I'm curious how The NYTimes got tipped off to this story before Matt Cutts and the Spam Team. Since this had been going on for month, I wonder how long The NYtimes had been investigating it. The article says, "Last week, The Times sent Google the evidence it had collected about the links to JCPenney.com."

Where is the boundary beyond which the link-building is considered black hat? I remember in the past black hat was referring mostly to deceptive tactics like invisible links, serving different page to Google and to humans for the same URL, etc.

Our quality guidelines are here: http://www.google.com/support/webmasters/bin/answer.py?hl=en...

We've talked a lot about why buying/selling links that pass PageRank violates our guidelines. The best starting point is probably this blog post: http://googlewebmastercentral.blogspot.com/2007/12/informati...

Blatantly off-topic, but...

Matt, you are everywhere. I'm amazed that you have time to actually do work, let alone sleep, considering how prolific you are on blogs and forums.

It gives me the Warm Fuzzies™ to know that an actual human being is indefatigably fighting search spam 24/7. On behalf of teh interweb, thank you.

Thanks; I get to work with a wonderful/smart group of people at Google--they're doing the really hard work. Both webspam and search quality folks are crunching on linkspam and content farms, so I hope we'll have more to talk when some of the things in the pipeline make into production.

So Matt, is this link profile not complete spam?

http://www.opensiteexplorer.org/www.couponover.com%252Fcoupo...

Yet this page ranks #1 for "chegg coupons" weeks after reported as spam in GWT. For every one JC Penney there are 100 "couponovers." Sickening.

Glad to see the folks at SEOMoz getting a hat tip, albeit indirectly. This is the tool that the researcher used to discover JCP's links:

http://www.opensiteexplorer.org

I thought Yahoo site explorer shows more links than just 1,000?

How long until someone pays for blatant black-hat SEO on a rival's site in order to get it trashed in the google ratings? How would you be able to tell whether that's happened?

For anyone who wants to take a look at the links heading to the site: http://preview.tinyurl.com/jcpseo

So if JCPenney sells black hats.. shouldn't they rank #1 for black hat?

This story should be more about Google's ranking code than a none story about JC Penny paying a SEO company, who did their job.

Matt_Cutts, people pay 3rd partys to "get me to the top for xx" when they deliver, you penalize then into oblivion?

Yes, Google does generally penalize that behavior, if the SEO companies cheat. Do what JC Penney apparently did -- use a load of unrelated sites to artificially boost its standing in search results -- and Google has a good reason to penalize you. Google can only stay trustworthy (and therefore relevant) if it eschews results like that.

Reply on news.ycombinator.com