Matt Cutts and Amit Singhal Share Insider Detail on Panda Update

[tedster]

Senior member g1smd pointed out this link in another thread - and it's a juicy one. The Panda That Hates Farms [wired.com]

Wired Magazine interviewed both Matt Cutts and Amit Singhal and in the process got some helpful insight into the Farm Update. I note that some of the speculation we've had at WebmasterWorld is confirmed:

Outside quality raters were involved at the beginning

...we used our standard evaluation system that we've developed, where we basically sent out documents to outside testers. Then we asked the raters questions like: "Would you be comfortable giving this site your credit card? Would you be comfortable giving medicine prescribed by this site to your kids?"

Excessive ads were part of the early definition

There was an engineer who came up with a rigorous set of questions, everything from. "Do you consider this site to be authoritative? Would it be okay if this was in a magazine? Does this site have excessive ads?"

The update is algorithmic, not manual

...we actually came up with a classifier to say, okay, IRS or Wikipedia or New York Times is over on this side, and the low-quality sites are over on this side. And you can really see mathematical reasons.

[Jane_Doe]

I do not think they will roll this change back. They succeeded in getting rid of many of the content farm pages. I would expect some tweaks but I would be very surprised if this change got rolled back completely.

[TheMadScientist]

Yes, everyone tends to think their site is better than all other sites but simply moving from position 14 to 130 is not something we can accept. Probably moving down to position 30 would be acceptable.

Traffic is probably close to the same between 30 and 130, in fact, according to the numbers I've seen, if you're not in the top 3 you're not getting much and if you're not on the 1st page all you get is a 'trickle', so why does it matter to everyone so much, except to tell friends where they rank?

Really, basically, if you're not on the 1st page you're not making any money or getting any traffic anyway, so what's the big deal, except bragging rights? I could be missing something, but I really don't get it, and they usually make adjustments to their rollouts to improve the smaller niches ... In fact, I wouldn't be surprised if there's one on the way soon (meaning in the next few days) ... There was an 'interesting' spidering cycle I saw a few days before the rollout (4 or 5 possibly as long as a week) and I saw it again a couple of days ago, so it looks like they're getting ready to add a layer like they were saying they would be doing from where I'm sitting.

[zoltan]

This was just one main term targeting the homepage. But the problem is that we lost at least 5 (Europe related terms) top positions (1-4), they were moved to page 2 (except one moved to position 9), and here is where we lost US traffic.

[zoltan]

And what is really strange is that according to Quantcast, in March 7 we had our second best day of overall traffic. For a moment I thought something improved (from my point of view) at Google US... Too bad that -although Canada, Italy and Spain had their best day ever in traffic- we are down 50% in US.

[zerillos]

Somoething happened on march 8. traffic down by another 10%

[frank72]

my traffic started dropping on the 2nd of March and the negative trend is still going on....when will the algo seattle down? Should I expect to see some up and down or....only down, negative trend...I am clueless.

[nuthin]

I am clueless.

Your not clueless. Let the update play out. Don't do anything too major or too drastic until things have died down a bit..

Short term solution during the update, there is always Adwords (yes shoot me).

[Shaddows]

Also, keep records of who is up and who is down during the waves of updates. Figure out who got rewarded at each stage, and try and find commonalities after the event. Then you can come up with a recovery strategy without constantly fiddling with pages hoping for an improvement.

Differentials are the key to understanding updates.

[crobb305]

We were spanked today, 60% drop in traffic. The site is not a "farm", and in fact is only about 110 pages, of well-written content for our users. We saw a spike in spidering yesterday (which we see every two to three months), then a huge drop in rankings this morning. I am unsure about how long to wait to file the infamous "reinclusion request". I am hopeful that I am just another (temporary?) victim of this algorithm and that it will sort itself out in time. Would you guys wait, or go ahead with a reincl. request? I know some Google employees were talking about a "fix" coming. Has this fix already happened?

[outland88]

What I'm seeing today is something totally different from the past two weeks. It breaks the mold. I'm seeing sites that didn't even vaguely step over the line totally vanish.

Crobb in the past two weeks did you make even minute changes?

[crobb305]

I have made no changes to my site in over a month. This site is 6 years old and we have added just one new article this year (back in January). We don't add new content frequently because we are not a blog (or a "farm"). But we do aggressively seek those who steal our content via DMCA. Therefore, our content is unduplicated. There is nothing on-site that should have triggered the penalty. This is all their silly algorithm changes.

[outland88]

Its fairly obvious where I’m seeing the problems right now and its on major keywords. I looked at five sites of competitors, who couldn’t even be vaguely accused of anything, and they had completely vanished. Would I imitate the replacements? Yeah, if I wanted to be broke.

[frank72]

I am seeing similar results today. I got a severe 30% drop in traffic today...out-of-the-blue. In my case it looks like i lost large portions of my long tail as the keywords I am monitoring are still there.

I know some Google employees were talking about a "fix" coming. Has this fix already happened?

where did you read this?

[walkman]

I know some Google employees were talking about a "fix" coming. Has this fix already happened?


where did you read this?

People that wrote about it labeled it as a fix, but Google said we'll be looking at hit sites and try to refine the algo. A more diplomatic statement

[crobb305]

I know some Google employees were talking about a "fix" coming. Has this fix already happened?


where did you read this?

frank72, there is a link to this on the main page here at Webmasterworld...and here is the discussion
[webmasterworld.com...]

People that wrote about it labeled it as a fix, but Google said we'll be looking at hit sites and try to refine the algo. A more diplomatic statement

I agree that is a more diplomatic statement. It's going to take a loooong time for any improvements to come down if they are going to take each complaint "into consideration" when looking at the algorithm, as opposed to offering a manual solution. Meanwhile, the algorithm spreads, more sites are getting hurt, and the saga will continue for months seems. I never expected to take a hit, some 10 days after "a fix" was said to be in the works.

[frank72]

I never expected to take a hit, some 10 days after "a fix" was said to be in the works.

ditto

for me it worked like this too...when the 24th of feb came the algo change left me untouched only to hit me 10 days later and the drop is still going on...

[Content_ed]

In my chasing after duplicate content the past two days, I've come across three of the top ten gainers, Amazon, eHow and Answers.Yahoo, hosting blatant copyright infringements of our work pasted by their community members. In all three cases, these infringements now rank above my pages in Google, or even force my pages into the "additional results."

It's getting very discouraging as my more popular pages are averaging sixty to eighty Google listed infringements. The only fun I'm having is searching on my own typos to find the copies:-)

[proboscis]

I just searched for a snippet from my home page and noticed several sites in front of mine, but they're not scrapers they are legitimate sites that are linking back to me. They just used the copied text to describe the link.

I would think Google would put my page first as the original source of the content since the other sites are linking back to me.

[tedster]

I'm seeing a lot more of that kind of chaos lately as well. The "original source" update was just a couple weeks before Panda, and it looks like something went wildly wrong to me.

[proboscis]

Thanks tedster, at least I know I'm not the only one, I did notice a few of my pages were affected around Feb 11-12 maybe that was the original source thing, then the rest went with Panda.

[Bewenched]

I'm seeing a lot more of that kind of chaos lately as well. The "original source" update was just a couple weeks before Panda, and it looks like something went wildly wrong to me.

The two weeks just before Panda we had some great traffic. After Panda, down, way down. Is that what you were seeing Tedster?

[dickbaker]

I've been re-writing pages on my site, all the while trying to figure out what it is about my site that caused my rankings to drop. As I go over pages of products from various manufacturers, I can tell which text is from the manufacturers' descriptions and which is text that I wrote fresh or re-wrote based upon the manufacturers' text.

I've been thinking all along that the similarity of my text to that of the manufacturers (and obviously other sites that feature the same products) could be one of the primary reasons for my rankings drop.

I did a search for a brand name product in a niche close to mine, but not my exact niche. I found the manufacturer in the first four spots, of course, then found four ecommerce sites above those. The fifth result was an ecommerce site selling that brand of widget. I looked at the product description for the model XYZ widget, then looked on the manufacturer's site for the model XYZ. The description was 100% identical, right down to the bullet point specifications.

I looked at the product descriptions on the other three ecommerce sites on page one, and two had the same identical description. The one site that had completely original content for the description was at #10.

I did a search for the first sentence of the description in quotes, and probably found all of the tens of thousands of sites that sell this manufacturer's products.

I'm not going to stop assuming that the similarity of many of my product descriptions isn't part of the cause of my re-ranking, as my site is only 10% ecommerce. I may not be getting the pass that 100% selling sites do.

I wish I'd found all original content on the first page for the model XYZ instead of duplication. Now I have to keep examining my site to figure out what's wrong, even as I play it safe and re-write content.

[tedster]

The sites I work with didn't have Panda trouble. Some lost a bit of Google traffic but it wasn't really appropriate traffic anyway - almost never converted. So 20 keywords were lost and conversions stayed the same. That sounds like an improvement to me.

What I was referring to is seeing scraped versions, or syndicated versions of content outranking the original source more often since Panda. In fact, a LOT more often.

The Scraper Update occurred right at the end of January. But rather than doing a better job of identifying the original source page, it demoted entire SITES that have lots of republished content. This happened three weeks or so before Panda.

Since Panda, for some reason it seems like the original source gets outranked by a copy more often. That's just anecdotal, I haven't done a rigorous analysis.

So the Scraper update, especially after Panda entered the mix, seems to be more about an entire site republishing content in mass quantity - but it's doing a poor job of knowing where the original source page actually is.

[hyperkik]

In my chasing after duplicate content the past two days, I've come across three of the top ten gainers, Amazon, eHow and Answers.Yahoo, hosting blatant copyright infringements of our work pasted by their community members. In all three cases, these infringements now rank above my pages in Google, or even force my pages into the "additional results."

I concur that with these changes you can find your original content pushed out of the rankings, unless you "repeat the search with the omitted results included". I've seen original pages that still show up in a top spot when you click to include omitted results, but don't otherwise appear anywhere in the search results. There may be an algorithm change as well - it's possible that when it's displayed in the results the page has shifted down a spot or two from where it was before a re-ranking, but Google seems to be imposing an additional penalty.

I think Google's ability to determine the original source got considerably worse with Panda.

[Content_ed]

@proboscis

Check their html to see if their links back to you are NOFOLLOWED. They usually are on such "tribute" scrapers.

If you want to find full infringements, it's usually best to choose a phrase from halfway down your page - eliminates the simple scrapers who automatically grab the first few paragraphs, so most of the results are full-blown infringements.

[Shaddows]

I think Google's ability to determine the original source got considerably worse with Panda.

I don't think they are trying to determine the original source, just get rid of sites which don't add value to the content, or do not make unique use of it.

If you want to protect your unique content, you have legal avenues available to you. Google is not it. All Google wants to block are valueless regurgitators.

[tedster]

Do you mean that the Panda Update isn't aimed at finding the original attribution (I agree) or Google Search as a whole doesn't have that goal? That second idea is worth more discussion, I'd say.

[walkman]

Shaddows, they have tried that many times. This time however they have also used (among other criteria) site appearance and the likes to nuke many useful sites.

[tedster]

Seems to me, without compiling a lot of data so far, that original attribution is only one factor Google tries to use. Especially when the original gets quoted on more a authoritative site, Google might well find their users prefer the "brand name" source.

[Shaddows]

@tedster- Google have made a half-hearted attempt at crediting the original source in the past. The Pre-Panda SCRAPER update was a shift in focus, IMHO. They stopped any attempt to publish the original- merely to reject the totally worthless. Which makes sense in today's plagiaristic SM world, from their point of view. Serve the searcher, not the content creator.

@walkman. Sure there is some collateral damage, and Panda massively muddied the waters. But the sudden appearence of republished content above the original, while suppressing the total crap, is indicitive of a successful change of focus.
------

In my opinion, Scraper was hugely successful. Hugely. I know most disagree. But if you shift your mindset to consider only VALUE, and not ORIGINALITY, it becomes a lot less... random.