Google Tweaks Algorithm; EHow Traffic Plummets

jfruhlinger writes "For some time there's been rumbling that Google's search results have been gummed up by low-quality pages from 'content farms,' written at low or no cost specifically to score high on common Google queries. Now it looks like the latest update to Google's search algorithm is having an effect, cutting into traffic to eHow (and cutting down the stock price of eHow's owner, Demand Media, in the process)."

Comments

and nothing

[Anonymous Coward]

of value was lost!

Re:and nothing

[Richard_at_work]

Demand Media tend to buy up a lot of pre-existing sites as well - Airliners.net, a fantastic aviation enthusiasts site, was bought by Demand Media in 2007 and rapidly went down hill resulting in a lot of members leaving :( Something of value is definitely lost once Demand Media get involved.

Re:and nothing

[Seumas]

Now, if they'd only start to drop their affiliation with that fraudster who runs Rip Off Reports. The guy who allows anyone to anonymously complain about any company or even any *individual* with private personal information and no validity to their complaints (say, they're just a bitter ex) and will only address the issue if you pay money to be part of the service as a "business" (thousands of dollars, if I recall). And, somehow, they are always magically weighted to the top of Google. (You can do a search for plenty of well known tech personalities and others who have had problems with this guy's blackmail service hiding as a legitimate consumer advocacy service).

Yahoo and others have not weighted them the same way that Google has prioritized the results.

Wouldn't it be a lot simpler

[vlm]

Wouldn't it be a lot simpler to just block all robot traffic to expertsexchange, ehow, and the like? Or even more trivial, allow users individual profiles to block specific user selectable domains?

Re:Wouldn't it be a lot simpler

[rgviza]

they buy up domains, prop up content farms on them, and all links on the content farms go back to eHow. So it's not as simple as blocking robot traffic to the offending domains (ehow and the like), you need to also crawl the content farms (easy because the SEO people register the pages to be crawled and google is already crawling/ranking them). Pages on some random domain that only have links to eHow, which are not on eHow.com, are probably farm content. Ditto for other farm content pointing to other domains.

The trick is for the robots to be sophisticated enough to ignore hidden (to user's browsers) content on the pages, look at the visible links and make a decision as to whether or not the content is legitimate based on where the visible links go.

Of course the SEO people will be on top of this and figure out a way around the algorithms. They always do...

Re:Wouldn't it be a lot simpler

[_0xd0ad]

Because 99% of the time, expert-sexchange just blatantly copied the question and answer from another website, and they try to hide the answer to get you to pay to register to see it. When someone wants me to pay to get something that I could get for free elsewhere, my basic reaction is "no, fuck you, you greedy bastard."

Sort of like the MLB wants me to pay to listen to live baseball games online while they're being broadcast live on free AM radio... and I can't even go to the AM radio station's website and listen to their live "on-air" streaming audio. No, fuck you, you greedy bastards... if I wanted to hear them that badly I'd go buy a cheap portable AM radio - it'd be cheaper and last longer than a subscription to your website. I only care to listen to one team's games anyway. As an added bonus, the radio wouldn't consume available bandwidth on my internet connection.

Well duh the stock fell

[elrous0]

If your company's business plan focuses exclusively (or even primarily) on gaming Google search results, then anyone dumb enough to invest in you *deserves* to be screwed.

Already done.

[kcbnac]

Previous Slashdot discussion:
http://tech.slashdot.org/story/11/03/11/1711252/Google-Introduces-Domain-Blocking-To-Search

Article discussing how:
http://googleblog.blogspot.com/2011/03/hide-sites-to-find-more-of-what-you.html

Re:Already done.

[vlm]

Interesting.. I did not realize that feature was available, because it is not shown to me in my results. There is no "block" option next to the "cached" and "similar" as seen in the blog posts etc. I do see in my search options that I can manually block up to 500 domains from search results. Nothing weird here, just a normal google account and a normal firefox with no unusual addons/extensions.

Maybe that option only appears for certain "special" domain names, or "special" searches?

Re:Already done.

[Maximum Prophet]

You have to be logged into a google service, click on a link in the search results, go back, then just that link will have the "Block" button. It took me a few tries before it worked for me.

Even better, you can go right to http://www.google.com/reviews/t to set things up in bulk. Then, when you search, at the bottom of the page should be a link like "Some items were blocked, click to see"

Re:Already done.

[Maximum Prophet]

You can go right to http://www.google.com/reviews/t to set things up in bulk. You do have to be logged into a google service first, and for the blocking to work.

Re:Already done.

[LizardKing]

You do realize that on EE, you can scroll to the bottom of the page for the incorrect answer?

FTFY

Re:Already done.

[demonbug]

Awesome. Experts exchange now blocked! This has immediately become my favorite google search feature.

I always read the URL as Expert Sex Change, so have never visited... So you're saying it isn't worth visiting anyway?

Re:Already done.

[onepoint]

Well google does cover that, just in case you do read Klingon, they want everyone to be able to read it http://www.google.com/webhp?hl=xx-klingon

Good riddance, Big Resource

[bhunachchicken]

I don't know about anyone else, but I was beginning to get very pissed off with looking up things on Google and constantly being linked to Big Resource, which was just a huge page of nothing.

Gettin' even bigger? Get as big as you like, you'll soon not see any visits from me...

Hit me badly too

[zakkie]

I run a website that is entirely my own work, is the result of years of research and involves many hours a day of new research. I am able to provide the data I collate for free to everyone because AdSense income covered hosting costs and allowed me to pay rent and buy food. I was not making vast sums of money, but I could do what I love and provide a useful resource to thousands of others. Now, scraper sites get ranked above me and even sites that cite me as the source rank higher than I do for many keywords. It's unfortunate, but for me this means less time doing actual original research and more time having to go out and market myself.

As a one man organisation, it's going to be really tough to keep going. I think Google have made a massive error here - by saying they can gauge the quality of a website (and its usefulness) algorithmically is arrogant and short-sighted. I hope they figure this out quickly. I really do hate having to sell stuff, even my own work!

Re:Hit me badly too

[drinkypoo]

I think Google have made a massive error here - by saying they can gauge the quality of a website (and its usefulness) algorithmically is arrogant and short-sighted. I hope they figure this out quickly. I really do hate having to sell stuff, even my own work!

You have it entirely backwards. Google has made the only intelligent decision here, by saying that they cannot possibly gauge the quality of all websites manually, and sticking to their guns about doing it programatically. That way, suing them over your position in the rankings is much more difficult because they can prove a lack of favoritism.

Re:Hit me badly too

[blue_adept]

Wiping out small independent publishers who "collate" mountains of information that they don't own the copyright for may well be what google intends to do. Let's be clear here. Are you an author? Do you create content? Or do you amalgamate other people's work, with or without their permission and/or using "fair use" provisions of copyright law? You don't really say, but I'm guessing from the tone of your post that you don't in fact create content. So, if your site is useful, provide the link, so we can have a better informed discussion about the merits of your site wrt the recent change in google's algorithm.

Re:Hit me badly too

[vlm]

So what's your site? ... What have you got to lose?

Zakkie links to http://www.carfolio.com/ on his /. profile page. I'm personally completely uninterested in the topic, but it looks like a real site as opposed to a content farm.

He's probably worried about losing his anonymity, knowing that /. is the most likely place on the net to have us all check his whois and reverse DNS records just for fun, etc.

Re:Hit me badly too

[gstoddart]

I think Google have made a massive error here - by saying they can gauge the quality of a website (and its usefulness) algorithmically is arrogant and short-sighted.

WTF? Google has always 'algorithmically gauged the quality and usefulness' of a website. That's what they do. That's what they've always done, and that's what search engines are for.

When they first came out, they were the best search engine because they explicitly pruned out the cruft and the link farms that had polluted all of the other search engines and made them useless. When I found Google, Yahoo and every other search engine got pretty much dropped, because Google actually returned content.

Google has become a multi-billion dollar concern by showing the 'arrogance' of thinking they can get rid of the shit and highlight the useful stuff.

If you truly have original research, and people actually cite you, then hopefully Google's new changes will actually help you and hinder the link farms and scrapers.

But, thinking it "arrogant and short-sighted" for Google to continue to do what they've always done is, well, arrogant and short-sighted. Google has been doing exactly this for over a decade, and continuously trying to make it better. To say that now they shouldn't be doing it because it might lower your page hits is hypocritical.

Personally, I'm all in favor of anything which gets rid of some of the useless damned meta-link sites that don't have anything of value, but exist purely to drive traffic from Google. I don't think such sites should be rewarded. In fact, I think they should be dropped from the search results altogether, which is hopefully what Google is doing here ... dropping eHow is a good start.

Re:Hit me badly too

[zakkie]

Well yes, but Google should not be judging a site's quality as such, they should be indexing the web. There is significant overlap in those concepts, sure, but currently their algorithm for indexing is broken - demonstrably so - and they are trying to do too much too soon.

I'm confused; are you in the index or not? If you're in the index, then your complaint is really about the ranking you're being assigned. But you just said you don't want them to judge quality (though earlier you argued that your quality is what differentiates you from your competition). How else should they be judging your content?

I am in the index, and Google is going a really shitty job of figuring out quality on whatever these new algorithms consider to be signs of quality. My site is older (they can verify that easily enough) than the scraper sites, has more links to it (the basis of the original page ranking system) and is also quite clearly being duplicated by these sites. Had Google stuck to just indexing the web as they did, the natural order of things would have been more favourable to me. Which IMHO is what it should be in this particular case.

Re:Hit me badly too

[bill_mcgonigle]

On my netbook, when I navigate to your site I see two pictures, a couple menus and a bunch of ads on the first screen. I have to scroll down to the second page to even get an idea about what kind of content your site offers. That's very atypical for a good website.

Re:Hit me badly too

[swillden]

Can you provide a sample search that shows this problem? The intent of the change was clearly to do the exact opposite of the effect you're describing. It might be useful for someone at Google to look into why their work has backfired in your case.

Re:Hit me badly too

[Draknor]

As for site layout, hey I did the best I can. I'm no designer, nor do I have resources to hire one. I'm not happy to ask for favours either, so I wasn't about to lean on mates who were more "designery" than me ;-)

Understood, but if you are "no designer", have no resources to hire one, and are "not happy to ask for favours", then why do you complain that sites that are better designed are ranked higher than you? I went to some of the copy-cat websites you listed and frankly, their designs are better (at least on the homepage -- I didn't spend THAT much time on them). You may have better content & all the blood, sweat, and tears put into compiling the data, but you still have to present it in meaningful way.

Your front page looks like an abandoned blog; ads on the top, a post from Feb 2011 and then some 2010 posts, and some random search links and car factoids on the right side. In a 3 second glance at your site, I have NO IDEA what this site is about. Some of your copy-cats -- bam, first glance, I know exactly why this web page exists.

Don't take this as criticism about you personally or endorsement of copy-cat behavior (its neither) -- just constructive feedback that your website is not doing your passion justice in terms of presentation, and that some design revamping & appropriate SEO may reap some substantial dividends.

Re:Hit me badly too

[jwhitener]

If I search for automobile technical specification you are the number one hit.
If I search for car technical specification you are the number two hit.

It seems that google is placing you fairly high for certain terms. Perhaps you lost rank when people search for shorter/less specific terms, like 'car spec', but that isn't so much an issue with google, as it is an issue with how your site is designed, how many others are linking to you, how they are describing the links, etc.. seo stuff.

Given that you seem to score really high for specific searches that are related to your content, I'm pretty certain that you can get higher in the ranks for the less specific searches with a little reworking.

Expert Sex Change

[HelioWalton]

They definitely need to tweak it further to get rid of or decrease the number of results from expert sex change.

like many programmers

[circletimessquare]

i've gotten experts exchange results in google searches forever, and i loathe them

however, not once, in years, have i seen "experts exchange" written in such a way in your post that it makes me think "expert sex change"

so thanks. thanks a lot. for making a bane of my existence somehow even worse. because now i will never look at "experts exchange" in google results again without seeing "expert sex change"

Re:like many programmers

[jon787]

Fun Fact: The MS Exchange team thought this was hilarious and named their blog msexchangeteam.com

Good EHow sucks anyway

[JohnRoss1968]

You can get better advise from the crazy drunk down at the park.

Was getting useless

[Quiet_Desperation]

I do some fairly technically oriented searches at work, and sometimes the first three pages of hits would be [1] sites that sell (or make you register for) copies of otherwise freely available documentation and [2] pages that are just random titles and snippets of other works without links.

Or there's some paragraph on a message board or in an article that has all the key words, but is useless, and all I get is 50 copies of the same article or posting. Some message board sites seem to be just copies of other sites with different CSS skins.

Re:Was getting useless

[mr_snarf]

My favourites are Google results pointing to a forum thread where someone is asking exactly what I want to know, and the only answer is 'google it'.

Judging Site Quality is why Google's Important

[billstewart]

The reason that Google is important is that they have good algorithms for judging site quality and showing the interesting relevant sites first. They became the dominant search player because PageRank produced better results than many other search engines when they started, as well as being fast and uncluttered. (DEC's Altavista, the original dominant player, was also fast and uncluttered, but Google's result quality was a lot better.)

If they weren't judging site quality, AdSense wouldn't be producing enough revenue for you to live on either.

natural order

[Runaway1956]

"the natural order of things"

Who wants to be the first to tell Zakkie that in "the natural order of things" there would be no intartubez? The internet itself is an artifact, and everything about it is artificial. There IS no "natural order".

So, what you are saying is, using some of Google's older models, you were treated well, and you were happy. With the updated algorithms, you are not being treated as well, and you resent it. This has nothing to do with any "natural order" at all. You simply prefer one algorithm over another.