Mercury News Interviews Matt Cutts - "Panda update working as intended"

[tristanperry]

Q How well is the new algorithm working, in Google's view?

A This change was designed to surface higher-quality content. We've heard from many publishers who are grateful, because their traffic has gone up after this most recent algorithmic update. I got an email from a user, and she said, "A couple months ago, I was searching for information about pediatric multiple sclerosis, and previously, low-quality sites were ranking above government sites." And she said, "I searched recently, and the government sites were ranking first." And that's the kind of feedback we like to hear from our users, and we've been hearing very positive feedback.

- [mercurynews.com...]

[snickles121]

Ya, but did he say how many complaints they got? I bet not.

[indyank]

when you search for "pediatric multiple sclerosis", most of the first page results are the same and they have just interchaged a few positions. I compared this with what was there before.

Panda has done more damage to a lot of other keywords but he seem to be repeatedly taking this as example, where the change is almost negligible.

[Planet13]

Panda has done more damage to a lot of other keywords...

Could you define "damage" a little more clearly?

Have the pages that suffered really been replaced by pages that were WORSE than those that had been ranking well? Or were they just DIFFERENT (but maybe less SEO'd).

In the few limited niche verticals that I have looked at, the pages that moved up were all of equal or better quality than the ones that moved down - I say this even in cases where a few of MY pages moved down. The pages that replaced them were, quite frankly, better - and my pages were GOOD, so the pages that replaced them were great.

Just I thought that because I had done better on page SEO that they should still rank higher.

Anyway, I am just trying to say that there were winners as well as losers. And yes, some scum floater to the surface, but I have seen a lot of other quality stuff taking advantage of the rising tide as well.

Just my two cents.

[Reno]

Again, his continued use of the phrase: "higher-quality content", with NO indication of what that means.

Come on Matt, say something USEFUL ~ if you truly want "to make things better for users", then define your terms so we all clearly know what you're talking about.

Spam is most definitely a problem and I'm the first to acknowledge they have a real challenge trying to deliver clean information, but he's not helping his company, his users, or the world-wide webmaster community by being so circumspect.

I'm waiting for the interviewer who is aware enough to follow up with the obvious question:

"Well Matt, you've gone to great lengths to say that this change was designed to 'surface higher-quality content'. You say it in almost every interview, but quality means different things to different people, so tell us exactly what you mean by higher quality?"

If he answers it honestly, without all the spin, we may actually learn something.

................

[chrisv1963]

Since when is the stolen re-written content on eHow "higher-quality content"? You see eHow all over the place since Panda.

[crobb305]

pediatric multiple sclerosis...previously, low-quality sites were ranking above government sites." And she said, "I searched recently, and the government sites were ranking first." And that's the kind of feedback we like to hear from our users, and we've been hearing very positive feedback.

Government sites rank highest on a medical search, and they are happy about that? What about university or private-sector research? What about real doctors? Give me a break. I can understand searches that pertain to federal regulations, but not about medical research. Ugh. It's also sad that the user who provided that feedback has been brainwashed/manipulated into believing that her government always knows best. lol. Again I say UGH.

[edited by: crobb305 at 6:56 pm (utc) on Apr 9, 2011]

[Planet13]

Since when is the stolen re-written content on eHow "higher-quality content"? You see eHow all over the place since Panda.

I think that ehow's increased rankings are a major blunder on google's part. I also see askmefast doing quite well in many searches, and askmefast might be the ONE web site that is actually worse than ehow...

But since I was just stating from my limited experience, and in those narrow verticals that I am looking at, there is no ehow presence.

So maybe I should knock on wood about that.

Either way, I just am reminded that the Chinese word for "Crisis" is made up of the words "Danger" and "Opportunity." I think we just need to look at the new opportunities that are created for us by the Panda update.

[MrFewkes]

< moved from another locations >

I wasnt hit by Panda - at all.

I just read this though from cutts, "This change was designed to surface higher-quality content"

I have read a lot about panda, people thinking things like "they were targetting our type of pages", "a penalty" - this sort of comment.

But if you look at cutts' statement - you could actually read into it that they have not ***looked*** at any type of low quality site at all.

They have looked for high quality (whatever that means for this algo) and REWARDED IT. Not penalised directly a site with low quality content by giving it a minus x.

I think this gives a lot of opportunity for us - I get the feeling that by looking at newly high riders - if you have spotted any - then their on-page SEO and on-site SEO and content hold the key.......

I think there is some good scope here for writing crawlers which assimilate data from various new rankers.

Just some thoughts really for those who may be struggling.

Cheers
Fewkes.

[edited by: tedster at 7:48 pm (utc) on Apr 9, 2011]

[tedster]

I'd say from the total of Panda communication from Google it is clear that they did BOTH - they clearly lowered sites that the algo tagged as shallow or poor quality and did not only promote what they measured as good quality.

[asabbia]

They have looked for high quality (whatever that means for this algo) and REWARDED IT. Not penalised directly a site with low quality content by giving it a minus x.

this is completly false

[TheMadScientist]

this is completly false

I think you're correct about your statement I quoted above being completely false ... Panda is not simply a penalty system, which means perceived higher quality gets a +, and perceived lower quality gets a -.

[edited by: TheMadScientist at 8:32 pm (utc) on Apr 9, 2011]

[crobb305]

Government sites rank highest on a medical search, and they are happy about that? What about university or private-sector research? What about real doctors?

I forgot to mention that in my industry a .gov user-profile page with a few spammy links their site is ranking on page 3 for a big term. It is clear that .gov got a boost.

[TheMadScientist]

Ya, but did he say how many complaints they got? I bet not.

Users (people in general) are waaaaay more likely to complain when they don't like something than to say when they do ... There have been many studies done on negative feedback v positive feedback ... One I remember said a person with a negative experience was likely to tell 10 people about the negative experience, where a person with a positive experience was likely to tell 1.

[Reno]

From MC... "And that's the kind of feedback we like to hear from our users, and we've been hearing very positive feedback."

OF COURSE it's the kind of feedback they "like to hear" ~ I'd like people to tell me that I look like George Clooney, but it wouldn't be true.

It's not what they "like" to hear ~ it's what they NEED to hear, and even taking into account TMS 10:1 negative to positive feedback studies, the fact is, Panda is being widely criticized at all levels, and there is much truth to those complaints.

So I hope the Google engineers are not high-fiving each other too much, as this update has (at this point) fallen short, and for many niches, it's an outright failure.

..................

[Robert Charlton]

pediatric multiple sclerosis

I'll take Matt at his word that this result has been improved, but this is a terrible example. In general, I am appalled by the kinds of sites I see giving medical advice, and content farms are among the worse.

What's most important to consider is the nature of the query.

It used to be that Google's medical results were stunningly good. If you search Google for the name of a condition only, the results still are impressive. Search for [widgetdisease] alone, by its technical name, or sometimes by a common popular name, and you will get some extremely good sites.

But as soon as you start getting vernacular or idiomatic, or start asking questions, the content farms start taking over the results. The farms are taking advantage of a loophole of sorts in the Google algo. Since the Mayday update, Google (to oversimplify) tends to consider longer phrases as more of a unit, and more specificity is now required to optimize for longtail.

The real authorities on medical subjects, though, very simply don't have the time or inclination to write a dozen different articles on the same topic in order to target every different variant of how someone might ask a question, and they use terms like "treatment" rather than "what do I do if I have?". As a result, in searching medical topics, you paradoxically often do better searching for the general topic rather than for specifics.

Matt's example of pediatric multiple sclerosis is rare enough that it's not liable to be competing with many "how can I tell if my child has..." type searches. It's surprising, though, how adding this preface to Matt's example disease trips Google up. (I'm avoiding the specific query, because I don't want to cloud the serps, but try Matt's example dropped into typical content farm type question formats). IMO, Google's algo is at war with itself here, and it's something they're going to have to sort out.

I should add that user generated content in medical areas is often dangerous. If you're lucky with UGC and if there's enough of it, you might get some contradictory advice on the same page, which will at least make you wary.

[Planet13]

So I hope the Google engineers are not high-fiving each other too much, as this update has (at this point) fallen short, and for many niches, it's an outright failure.

I don't have any clearer idea of what the team at google considers "quality."

Google is about as engineer-focused a business there is out there, and they do things "by the numbers." They will most certainly make any decisions based on the data they get.

I am thoroughly convinced they have a boatload full of metrics to help them determine whether their rankings are providing higher "quality" to visitors, or not.

I am sure they will be paying almost exclusive attention to what google users do - whether it is counting the visitors going to a site and then returning to google, measuring the clickthrough rate of SERPs, whether people are using the chrome block site extension more or less after panda, and whether more or less visitors are going to bing after viewing the google SERPs.

I think those will be some of the key metrics for google in determining if they have improved quality, and I would imagine that even with the huge daily search volume out there, it will be a while before they feel they have collected enough data and have been given enough time to crunch the numbers and draw any conclusions.

Whether you or I think that is quality is irrelevant to google. The opinions of webmasters and the press at large will count for nothing.

[Planet13]

I should add that user generated content in medical areas is often dangerous. If you're lucky with UGC and if there's enough of it, you might get some contradictory advice on the same page, which will at least make you wary.

What if some parents happen to follow the bad advice given in some of these "medical" sites and end up killing their child. I wonder if people will be able to sue google in the future because the site that provided that medical advice was ranked #1 in the SERPs.

[indyank]

It used to be that Google's medical results were stunningly good. If you search Google for the name of a condition only, the results still are impressive. Search for [widgetdisease] alone, by its technical name, or sometimes by a common popular name, and you will get some extremely good sites.

But as soon as you start getting vernacular or idiomatic, or start asking questions, the content farms start taking over the results. The farms are taking advantage of a loophole of sorts in the Google algo.

The real authorities on medical subjects, though, very simply don't have the time or inclination to write a dozen different articles on the same topic in order to target every different variant of how someone might ask a question, and they use terms like "treatment" rather than "what do I do if I have?".

I couldn't have expressed it better than that.

[GeraniumV]

Funny you should mention the medical aspect. I read an interesting story on the BBC where "Cancer Research" was throwing its experts at Wikipeidia to clean up there so called facts.
[bbc.co.uk...]

And Wikipedia is one of the higher quality sites. I had the same problem with my dad looking up a medical problem on Google and taking it to his doctors. He really believed the stuff he read (I forgot which site it was but probably eHow). I had to persuade him otherwise - by taking him to the expert sites- I think that's a real problem!

[Roaming Gnome]

We've heard from many publishers who are grateful, because their traffic has gone up after this most recent algorithmic update.

Of course they are happy. (Why wouldnt they be?)

My complaint is that keywordkeyword domains are still ranking high with no content.

[chrisv1963]

Search for "how to treat cancer of the pancreas" and you will find eHow (again) ranking first. This is a JOKE!
Or what about "how to avoid a stroke", ...
Ehow claims to have an answer to any question. We as webmasters know that the content is re-written, often by non-professionals and with false information. However, the average user doesn't know this and will always consider this as the correct information.

But, ... the Panda update is working as intended. The Google engineers are happy with the "higher quality" results, even when they send users to wrong information.

[norton j radstock]

Search for "how to treat cancer of the pancreas" and you will find eHow (again) ranking first.

From eHow, in short:
1. Watch for symptoms of pancreatic cancer to aid in early detection and treatment.
2 Get a second opinion from a doctor who specializes in cancer of the pancreas.
3 Undergo surgery to remove all or part of your pancreas.

Job done!

[onepointone]

I would trust them more for searches like "how to pick your nose" or "how to boil water"
They do have pages on those and do rank in the top 3.

Actually, I don't fault g totally for these things. If the site is better at seo, (and they whitelisted it) what can they do? :)

Plus the internet is great & all, but people rely on & trust it wayyyy to much IMO.

[chrisv1963]

I would trust them more for searches like "how to pick your nose" or "how to boil water"

And I would trust them for "how tow rewrite articles". They are an authority in this field!

eHow is the biggest content farm ever and Google puts them on the first page for almost any query with "how" in it.
Actually, Google should give them at least a -10 penalty for any query.

Why doesn't the algo detect this content farm? It shouldn't be that difficult. Trying to cover too many niches = content farm, unless you are a newspaper or recognized news source.

[TheMadScientist]

<Slightly/SortOf-OffTopic>
Well, I thought I was going to bed a couple of hours ago, but can't sleep, so here I am posting again, and it's late, and I may well be grouchy, but I'm sitting here thinking what's even a bigger joke than Google's rankings of eHow is a publicly traded company, like Demand Media, allowing the content on their site in the first place ... I think I saw somewhere earlier in this thread where someone wondered if Google could get sued for ranking bad advice someday, but I really think it's Demand Media's responsibility for allowing the asinine posts on their site in the first place ... Demand Media's existence in it's current state is a much bigger joke than Google's rankings, imo.

I wonder if they considered DaSpam Media as a name but decided on Demand, because DaSpam might be too much of a give away as to what they seem to be at the core?

Can we get an eAsinine Forum so I can say what I really think without getting all OT please? lol

Yeah, you can delete this one tedster, but I really get tired of listening to how bad Google is for ranking eHow, when the reality is Demand Media is responsible for eHow and no one seems to say anything about their responsibility for their garbage...

Did the CEO really say he was offended to be called a content farm? Personally, I'd be embarrassed to be CEO of a company with the absolute lack of quality and duplication (or near duplication) there seems to be in so many cases on eHow ... He did say they were going to clean things up though; I wonder if this means they'll be changing their advice on how to effectively pick your nose from the following to something else?

Take him to the taxidermy.

Have the little guy mounted on the wall, the chicks will dig it! Guaranteed! Boogers are always very impressive, even more so than any game you could hunt. Find a good spot for everyone to marvel at him.

http://www.ehow.com/how_5067366_effectively-pick-nose.html
(I can't nofollow a link to the source like they do, but I can post one in plain text.)

He was offended to be called a content farm?
Maybe he should have been enlightened...
</Slightly/SortOf-OffTopic>

[TheMadScientist]

And what shows how wrong we are about what quality is?
The 95,000+ FaceBook Likes on the page I cited above...

We're the idiots for thinking quality means 'mature' or 'adult' or 'well written' or 'correct' ... Quality is a Joke ... I think it's time for me to retire ... I can't write S*** like that eHow page...

[indyank]

and i see the ad for "Google Chrome" just below it. What a perfect match.

Google asks people to remove all bad content but doesn't expect e-how to do it.

This page that "TheMadScientist" has referred gets 35 likes and probably some good number of tweets! Google probably loves it for those likes and tweets more than the content on the page.

if google hasn't white-listed them, then the panda algorithm should be a genius for turning a blind eye to all this S****

[indyank]

Did you notice the related ads on the right sidebar.It actually leads to a page with no content but only google ads. There is this message.

About these related ads:

These advertisements are related to a specific topic which were automatically generated to match the content of the web page you previously viewed. The links are administered, sorted and maintained by a third party.

Amazing partnership.

[jinxed]

"How to Effectively Pick Your Nose"

I have now seen it all.