Panda 2.2 from two perspectives

[launch3]

I've breezed thru every Google update for the past 10 years.
None have affected me until panda.
The April panda update leveled my 3 websites (60%+ drop).
No recovery since.

However, I'm going to show you from a user perspective why Google is currently broken.

I just updated to Firefox 5.0.
My Google toolbar no longer worked.
So, I googled this exact phrase (i urge you to do the same as an example):

fix google toolbar for firefox 5

result #1: irrelevant
result #2: irrelevant
result #3: somewhat relevant
result #4: irrelevant
result #5: irrelevant
result #6: irrelevant
result #7: perfect
result #8: irrelevant
result #9: irrelevant
result #10: irrelevant

(page 2 search results weren't any better)

result #7 fixed my problem and my toolbar works. But why wasn't it #1? It was the MOST RELEVANT.

What do I take from this?

Your websites and my websites are currently sitting below irrelevant, outdated results & in many cases scraper and pathetic results.

This is simply the latest search I've been frustrated with.

Yes, I've fixed up my websites with a fresh new design, added Disqus comment system (which is working great), and other cool new features.

Yet, it makes no difference.

Why?

Google is broken and someone over there had better take responsibility and fix it soon.

I can no longer find what I am searching for with ease.

Great, helpful, relevant websites are no longer near or at the top of the results. It's a joke.

My advice, and the advice I have seen mentioned on occasion, is to sit tight. Clean up your websites, but realize the problem is not simply you or your fault. I've been around webmasterworld since inception and don't often post. This finally annoyed me enough to hoist me off the fence.

[tedster]

Your observations have been mirrored by many people. Even Google has acknowledged that the scraper site problem is out of hand.

With regard to the broken search you mentioned, please note that we do not usually post specific search terms here. So this thread is not an invitation for others to post examples - there are many broken searches, that point is clear.

[falsepositive]

I think a Google spokesperson admitted that Bing did better in some searches and that they need improvement in some other search areas. Supposedly they are doing something about it. It's great that they realize this and they are "on it". But it feels like such a long haul though. What is "long term" for them? Hopefully it doesn't take 3 years to fix these problems.

[mrmobility]

When did Firefox 5 launch and how old was the article that fixed it for you?

[indyank]

1) I am noticing the scraper problem ever since panda was released and I never saw it so widespread as now. Google was doing an excellent job before.

2) Firefox 5.0 being so new, you obviously wouldn't have many correct answers for error fixes. But if #7 is perfect, it should rank ahead of the irrelevant ones and definitely the scraped ones should disappear.

Coming back to the scraper issue, I think google had given up on the earlier method of detecting scrapers (duplicate content) to save resources, as they probably believed that the panda algos (run manually) will address them.But this one is so far a nightmare and no where close to the earlier one.

[tedster]

Relevance is not the same as accuracy. A 100% falsehood can still be relevant for a query. No algorithm today even tries to measure accuracy.

In addition, this query is confounded by numbers - Google doesn't do a good job with software version numbers unless you use quotes. Otherwise a page about Firefox 3.5 can easily be seen as relevant for an unquoted query for Firefox 5

[walkman]

The scraper is not easy to fix IMO.
Do you go sentence based, full text, paragraph, 3 consecutive words? Imagine what happens when the same sentence shows in in 100...100,000 sites? First it takes lots and lots of resources to crosscheck all the web. Second you may do even more harm.

As some have pointed out, scrapers might be surfacing because the owner may have been pandalised and the scraper has not, being too small and under the radar.

What is "long term" for them? Hopefully it doesn't take 3 years to fix these problems.

And that's our problem. Not much we can do now.

[Frost_Angel]

Launch3,

I am in the same exact boat as you:

"I've breezed thru every Google update for the past 10 years.
None have affected me until panda.
The April panda update leveled my 3 websites (60%+ drop).
No recovery since."

I could have written this about my 12 year old site. Honestly - I don't know what the issue is. I don't do weird black hat voodoo crap on my site. I write everything and it's in depth.

I redesigned my site, cleaned up all my pages, little advertising - I submitted a reinclusion earlier this month and a week later got their standard poopy letter that tells you nothing.

The sites that ARE ranking for my keyword phrases, total, oudated, MFA crap.

I am really frustrated. I don't know if they just set out such a huge net that there were innocent casualties...I just don't know? I don't feel like I'm a complete idiot - but this update is making me feel that way! LOL

I know some are seeing recovery after this recent 2.2 update that happened recently - but nothing has changed for my site. Around April 10th someone sank my battleship and I've been unable to resurrect it! LOL

I wish everyone luck - because hard work and quality content and playing by the rules won't get you jack!

[Jessica97]

@Launch3, that's been my take on Panda ever since it was implemented. Did Panda actually improve search results? For most of my industry and when I do any personal searching--I would have to say not. I won't post exact terms here but yesterday I needed to find out how to repair a certain item and the first 4 results were nothing but trash sites. The term wasn't about software, had no version numbers, and wouldn't be considered one that was ultra competitive. That part is the most frustrating.

And I'm certainly not saying that the scraper problem is easy to fix. But if the results that are returned are worse then when they started with Panda...well, I'll let each of you finish that sentence on your own.

[launch3]

mrmobility, i found out about firefox 5 today on webmasterworld and the search result that fixed it was posted yesterday on a mozilla thread about this very problem.

update: this thread has already taken the #7 search result position where my fix was located. the fix link dropped to #10 in the span of 1 hour. irrelevant results still take up the rest of the first page.

[mrmobility]

I was wondering as I think it's a tough query to use as a benchmark being so new. Bing for example doesn't seem do any better with it.

[launch3]

what i find interesting is that this very thread included the exact search phrase i used and it's on a very reputable website. it immediately got listed in google, yet it still only managed the exact #7 position which oddly enough replaced the fix link itself and moved that down to #10. it had no effect on the rest of the results. this updated algorithm courtesy of panda won't permit it a top spot despite the obvious exact phrase match, freshness and better relevancy than the other results.

[tedster]

Those new changes in rankings are not Panda in action - everything that happens cannot be looked at through Panda colored glasses.

However, what you noticed about position #7 is an interesting observation. Maybe there is something "special" about that spot right now, just as position 4 and position 11 seem to get special handing at times.

[walkman]

Jessica, launch, Frost... and others take my advice:
It does not pay to even make your views on Panda known. Google is happy with Panda based on whatever goal they set out to do. If not they are taking their sweet time to experiment live until they fix it. Since Feb they have even added new criteria to catch other sites. First thing they did is call the pandalised sites "low quality" and told you /me to write articles that show the other side of the story, make the site look like an iPad or whatever. That was two months after Panda 1, IIRC. If that seems odd or unhelpful to you /me, it is how it is. They make the rules.

Eventually here will come someone who call you a spammer and a cheat because his/her site didn't get pandalized and since someone said that they recovered you must really be a loser.

Maybe in 4-8 weeks you'll have better luck, if not wait for another 4-8 weeks...rinse and repeat. Maybe 2 months from now Matt Cutts will come and give us another 'hint,' the usual crowd will applaud them for being so communicative with the commoners and we'll spend the next 6 months trying to 'fix' whatever is supposedly broken. And then we'll have to wait 4-8 weeks for them to run the data...and you get the message and your first social security check.

Whether there's content changes or more at play, it has been like this for 4 months now. Eventually we need to get the message.

[launch3]

point taken tedster. taking off the panda glasses. but yes, there is a #7 spot intrigue.

i agree walkman. i closely followed your initial thread on this subject. i'm just going about my business as usual trying to make my sites an even better user experience. adding cool widgets this week for other webmasters to follow my content from their sites. one thing panda did was give me incentive to NOT rely on google for traffic anymore. i was complacent with already great sites and over 10 years of consistent traffic. heck of a wake up call, but heeded. if i build a better brand, people will continue to come back with or without google.

[Frost_Angel]

Launch3,

I am in the same mindset as you. I will never again rely on Google for traffic and actually not even using it anymore as my search engine. LOL

It was a wake up call. In the end I tell myself - I didn't even start this website for the money back in the late 90's! It was to help people...so there was no disappointment when money wasn't the goal. I am again going back to this way of thinking - unfortunately I have a mortgage and kids to feed now....so money was a nice perk - LOL

My Mom's advice - "Death and taxes....only thing you can count on." LOL

Thanks Mom.

[londrum]

i think that is a bit of an unfair test. I only got prompted to update Firefox today, and presuming that not everyone had the same problem with the toolbar, you have allowed next to no time for relevant pages to be written.

And even if some were written today, it is highly unlikely that they will have gathered many backlinks, or bookmarks (who would bother to bookmark that?). So how are google supposed to rate the page? its not like they can measure stuff like "time on site" or bounce rate, because the answer is likely just going to be a few sentences long... people will be in and out of the page in 20 secs.

as for the site at no.7, there is no way for google to "know" that an answer is correct. until they've gathered enough user signals, they can only rate the pages by freshness and trust. you are expecting too much.

[launch3]

hey londrum. could this test be a bit unfair? sure. i'll grant that. but this has been happening to me for several months now. wading thru results that obviously shouldn't be there is becoming more regular.

my last search in google the other day was for an exact title of an article of mine that's been popular for years. Now, above it was a scraper site with a sentence and a half of content. Literally, it cut off in mid second sentence on this person's site and yet it was now #1. had no relevance or point to it. it was a head scratcher.

when new books or products launched, google would know to place it atop the results immediately. now it isn't necessarily the case. the algorithm rules have gone berserk.

[londrum]

sure, i agree with you on that. i think its got a lot to do with freshness and trust.

a scraper site will ALWAYS trump you on freshness, because your site will obviously have the oldest copy. as for the other stuff... who knows how it works. havent got a clue. panda is a big mystery to me.

[tedster]

This whole scenario seems related to the outrageous jump in indexed URLs that Google went through when the Caffeine infrastructure came online. They have a big challenge in making sense of THAT much data and so far, they do seem to be struggling except on the most common queries.

I'm guessing that their solution is to meet the challenge rather than to back off on using all the data and use only a smaller subset. Bing seems to have chosen to keep their index significantly smaller, and for some queries, that does seem to give better results. But I'd rather see Google meet this challenge, no matter how big it may be.

[superclown2]

I'm now getting used to the idea of giving up on searches. I actually bought some books today for information on some subjects I've been researching because I just cannot afford the time it now takes to wade through panda droppings. Pre-panda I would never have dreamed that this could happen. This is progress?

I'm all in favour of Google moving forward but can't they let us have a search engine that works as well as it used to in the meanwhile whilst they make this thing fit for purpose?

[supercyberbob]

Again, good to know I'm not the only one who thinks some of these serps are super junk.

The big question now is if Panda really is an improvement.

Oh yeah, take off your Google goggles before attempting to answer the question.

[Crystal Pegasus]

I did a search a couple of weeks back, and 5 of the first page results were the same article, just on different sites! What's more, I checked the next 2 pages of results, and in total on the first 3 pages there were 17 of that duplicate article! How is that useful?!

And it wasn't even a good article, just pretty woeful MFA fare. Nor was it an obscure kind of search that there was unlikely to be many results for. In the rest of the results there were also a couple of duplicates for another 2 articles. Crazy and pretty useless.

[koan]

Again, good to know I'm not the only one who thinks some of these serps are super junk.

The serps are derps.

[tedster]

We're not going anywhere now - so this thread is locked.