Many Weeks since the Panda Update - Any Improvements?

[grimmer]

It has been 2 weeks now since google's Farmer update on Feb. 24th, for the sites that are affected, anyone see any improvements? For my site, we have started to remove low quality content a week ago, but have not seen any ranking improvements so far.

[crobb305]

I am seeing a complete rollback since this morning.

I sure wish I was seeing a rollback. I'm still down -58%. Maybe you have improved ranking? Or are you seeing other sites return as well?

I do see a rolling pattern in my hourly histogram today. I haven't seen that lately.

[kd454]

No rollback here, I do see slightly different serp's using different browsers.

The changes/improvements I have made to my site have had no effect, bounce rate has gone down even lower to around 28%.

I am thinking there is not going to be a quick rebound (if any) G would not make it that easy.

[outland88]

Google slides a bizzaro world type update every week into my areas that lasts about six hours. Its not related to the past or browsers.

If anything they need to roll it back to pre-Mayday. Lets not mess around I want the whole pie.

[hyperkik]

But they never defined what low quality is.To me it is those pages that don't get hits as good as other pages on your site.It is not about thin content really but it is more relative to other pages on your site, in terms of traffic.

Panda hit substantive pages on my site that have traditionally received very high traffic, while having no negative impact on traffic levels to the forum on the same website. That is, the thinnest content on the site is doing "just fine, thank you very much".

So, I'm wondering... What if the previous love from Google for our affected website was in significant part due to the sheer number of incoming links from this one site? If so, and this is something now taken into account in the Panda update, it seems reasonable that all those "votes of confidence" from a single source no longer weigh as much as they used to.

It's fair to speculate that part of this is Google re-assessing how it weighs links to sites - to a large degree all we can do is guess. But I don't recall any statements from Google suggesting that this was about links as opposed to content. I also don't see how a site like eHow would get a huge boost from such a change. Further, if I look at the evidence available to me, for me this has not been a site-wide phenomenon. Portions of the site with the best and longest-standing links have been much more affected (@40% drop) than forum pages to which there are few to no outside links. Forum traffic has, in fact, increased by about 15%.

[indyank]

hyperkik, I am talking about the ratio of low traffic pages to overall indexed pages, pulling down your best pages.

you will have to read it in full.

[hyperkik]

Read what in full? I read your posts in full, if that's what you mean. I just don't happen to agree with your conclusions based upon the evidence that is available to me.

Another point I should have made in terms of this being a re-weighting of links is that, were that the case, I would expect a much broader spectrum of how sites and pages were affected, not relatively uniform reports of traffic drop-off in the 40-60% range.

[indyank]

Though i don't mean traffic as the only factor used in it. It should read as "Quality" which is computed from all relevant signals including traffic.

relevance might have been determined based on a single page or other related pages on the site.

But it does look like google is computing "Quality" for the overall site and applying them on top of relevance signals for the individual pages.

1) Relevance is a measure of the page (pre panda)
2) quality is a measure of the site + the page. (post panda)

if the second one is applied on top of 1 and your site gets ranked poorly in quality, then your best pages do get pulled down.

[walkman]

I am seeing a complete rollback since this morning.

No rollback, just seeing SERPS from 1-2 weeks ago. Many saw an increase in traffic right before Panda started to chew them.

Does anyone who's been adversely affected have a large number of forum signature links showing in Webmaster Tools?

Maybe, but it's all Panda, 24/7, right now. The links might be fine (albeit ignored) and something else is hated by Panda.

1) Relevance is a measure of the page (pre panda)
2) quality is a measure of the site + the page. (post panda)

if the second one is applied on top of 1 and your site gets ranked poorly in quality, then your best pages do get pulled down.

That's exactly what I see. I cannot rank for pages that every sane person would agree that they are very good and I have a top domain name (decade plus online).

[Pjman]

I have seen nothing on my end to indicate that Feb. 24th Panda had any thing to do with Links. I would think that links would be re-valued based on the new site quality score, but I have not seen anything to indicate that either.

Also usually Matt Cutts is a straight shooter. I haven't heard him say anything, but quality (thin content).

Every site I saw hit, including mine, had thin content or unmoderated user generated content in there some where that account for greater than 5% of the pages of the total site. My two absolute authority sites, nearest competitor has 1/10 the content, were hit hard because of UGC. Everything was moved back 3-6 positions in the SERP. For most from Page 1 to Page 2. G referral traffic down 70%.

I never paid attention to that before because Pre-Panda, Google only ranked pages, not web sites. Now everything has to be clean to get the Google Gods approval.

Glad I use twitter and FB more now. That and my newsletter have saved me.

[Pjman]

Just thought I'd point this out for anyone trying to compare pre-Panda results to post-Panda. When you use a well known site that does cookie scraping google searching, in my niche at least, those results are all pre-Panda results verbatum.

[TheMadScientist]

Let me see if I can clarify what I keep trying to say and have been thinking for while in case it might be important to someone:

It's everything people keep saying rolled into one.

It's not like what we've seen before ... It's all the little things people keep talking about lumped together, which makes it very difficult to pin down what the cause is, because for each site / page it may be slightly or totally different.

[outland88]

It's everything people keep saying rolled into one.

No lie, Its like I told you a hodgepod. Nobody's wrong, nobody’s completely right.

I did figure out the bizarro world changes I have been reporting on. It has had me scratching my head for a while with stats. Google is actually running two distinct sets of results in my areas believe it or not. I catch the tail end of it dependant on the times I check in. But let me say the lab techs could be changing the user behavior experiments tomorrow so nothing is chiseled in granite.

To me it's literally amazing the lighting speed at which Google can kill business or kill a business. It's almost James Bond "take over the world" type stuff.

[rlange]

hyperkik wrote:
But I don't recall any statements from Google suggesting that this was about links as opposed to content. I also don't see how a site like eHow would get a huge boost from such a change.

You're right. It may be out of desperation that I'm thinking *too far* outside of the box.

Further, if I look at the evidence available to me, for me this has not been a site-wide phenomenon. Portions of the site with the best and longest-standing links have been much more affected (@40% drop) than forum pages to which there are few to no outside links. Forum traffic has, in fact, increased by about 15%.

Another point against my hypothesis is that, after a little more research, those 100k+ incoming links I mentioned were directed at only 14 of our ~2k pages. Two of those 14 had 40k+ links each. Looking at those two pages, they saw drops in traffic that were pretty much in line with the drop in traffic that the entire site is seeing. Basically, nothing out of the "ordinary".

Anyway, there's one section of our site that has always had the potential to be rather weak (business descriptions provided by the businesses themselves). There are a lot of pages in that section that dropped more 50 positions, but there are also quite a few that have risen more than 50 positions. Some that I thought were weak saw and increase while some that I thought were decent dropped. There appears to be no rhyme or reason.

[Pjman]

You're right. It may be out of desperation that I'm thinking *too far* outside of the box.

I think we are all grasping at our hair on this one.

The only reason I say it's not links is like you said, Google never said anything about links. I didn't hear chatter from anyone credible about links.

If history serves, they are always up front about what the new algorithms target.

Conspiracy Theory: They do have a new CEO; so maybe they are changing the communication culture too.

[Bewenched]

I haven't seen any significant change at all, and some days I swear it's worse.

[crobb305]

I am seeing improvement in my traffic and rankings. My traffic is up about 20% from last Wednesday, and down only about 40% from previous weeks (whereas last week I was down 60% to 65%). This is a positive sign.

[grimmer]

crobb305, can you clarify whether you have made any changes, and whether any of the changes may contribute to the improvement? Thanks.

[honestman]

Many people I know say they are frustrated that they get very, very specific results related to companies selling xyz widgets, as opposed to directories of such companies or a website which finds results which discuss the idea or items searched. In other words, more weight seems now to be given to narrow interpretations of a search term, which makes research much more difficult--and research is (was) the basis of the web, I thought. Just experience based on people I know and my own experience. Of course there are exceptions with the big name brands, etc. and your mileage may vary, but this is a pattern I see.

[tedster]

One correlation I am hearing about is the presence or absence of deep backlinks. This may not be cause and effect, but it makes sense that weak content would not attract the same kind of deep backlinks that good content would - so correlation might be strong in this area.

[walkman]

Yes Tedster but individual product pages are very hard to get links, Google know it. unless you're amazon

[tedster]

Agreed - that's one of the reasons I emphasized that some people are seeing a correlation, but it probably is not a cause-and-effect relationship. Since that's the case, going out to get deep backlinks would not fix the lower rankings.

[ken_b]

I am hearing about is the presence or absence of deep backlinks.

I suspect the source/quality of those backlinks to deep pages matter.

I was looking at my backlinks (via Yahoo Site Explorer) yesterday and was surprised to see how many of my backlinks went to inner pages.

Most of those links are what I'd call junk from scraper/aggregators/whatever and are certainly not links I ever went looking for, but there they are.

Meanwhile they certainly don't seem to have helped me in the Panda/Farm/Farmer update.

May not be the case for everyone, but seems to be for me.

(my backlinks: less than 2,000 to the home page, over 30,000 spread over 2,000+ inner pages, according to Y.S.E.)
.

[Jane_Doe]

I suspect the source/quality of those backlinks to deep pages matter.

You can beat out Wikipedia and the federal government sites even on very competitive one and two word topics with solid enough, deep backlinks. But otherwise, backlinks being equal or inferior, the other sites win out.

Meanwhile they certainly don't seem to have helped me in the Panda/Farm/Farmer update.

I think this go round the algorithm is more selective in the past on which backlinks are going to really give you a boost. I think that is why older sites often have gotten hit harder. They tend to have collected more scraper and easier type links over the years, but now those links aren't counting for as much.

My traffic is up a third on one site in a competitive area, I think in part because I have been very proactive in getting editorial links from other high ranking sites on the same topic. With the older sites I didn't have to do that (until now that is).

But hey, c'est la vie. I'll just have to do the same for the older sites now.

[Dan01]

I don't think it provides users better SERPS by penalizing sites (on a sitewide basis) because of some of their content. I don't know how many times I have typed in a query and Wikipedia popped up first. Sure Wikipedia has good content, but it didn't answer my question. I found the answer (hopefully) on page four or five from a lesser known site. I don't think sitewide penalties are a good thing.

[dickbaker]

I just found something interesting while trying to figure out which pages I should be re-writing. I went into Google Analytics and used Traffic Sources>Search Engines>Google, and then entered the main keyword for each brand of widgets I feature on the site.

I did this for 2/1/2011 through 2/22/2011, and then for 3/1/2011 through 3/22/2011. GA shows a synopsis that says that there were X number of visits from Y number of keyword combinations. I did the March period as a percentage of February for X and Y.

Most of the brands were showing ~30-35% for the X number of visits and maybe 40-45% for the Y number of keyword combinations.

A couple of brands really stuck out, though, with 54% of visits and 62% of keyword combinations, or even 77% of visits and 77% of keywords.

I took a look at the pages for those brands that had much higher percentages. They were brands I hadn't touched in a few years, as I didn't think they were that important. The text on the page was lifted 100% from the manufacturers' sites (I didn't remember even doing that, it was so long ago). I could take a sentence, search for it in quotes, and find dozens or even hundreds of other sites with at least a portion of the same text.

The other difference was that there was a lot of text, versus the other brands where there was maybe half as much text, but it was primarily written by me.

So, the 90% to 100% scraped content beats the original content because there's more content.

Reminds me of an old Woody Allen joke:

1st diner: "The food here is terrible."

2nd diner: "Yeah, and such small portions, too!"

[Dan01]

Dick, I think that makes sense. The longer articles / content pages / whatever can do better. They will contain more keywords. What doesn't make sense is that Panda was supposed to ding you for scraped content, but didn't.

[Dan01]

Shatner

Wrong in every possible way. First, Farmer & Panda are the same thing. Second, Google SAID Farmer/Panda penalized content farms, but that has proven to be just PR spin unsupported by the facts. In reality a greater percentage of content farms weren't penalized than any other type of site. In fact many content farms benefitted from Farmer/Panda.

Thanks for the correction. I think Dickbaker provides some evidence to that point right above.

[dickbaker]

I'd be tempted to just go to the sites of every manufacturer I feature on my site, copy their text verbatim, and paste it into my site, rather than write new text.

There has to be another shoe falling soon, so I'd better clean up those scraped pages. I'll just make sure to use a lot of words.

[Dan01]

Here is a question: Why do scraper sites sometimes rank higher. I have spent hours working on a particular piece of content, and then poof, someone scrapes it an ranks higher in the SERPS. It doesn't always happen like that, but when it does it hurts.

Here is why I think they can rank higher: Google says that their robots revisit sites more frequently if content is added quickly. If all I had to do was scrape content, I can get tons of content up quickly. If Google believes they were the first to post the content, they must have created it.

[edited by: Dan01 at 11:37 pm (utc) on Mar 23, 2011]

[Pjman]

In one hour, it has been exactly one month of being Pandalized, at least for me. Happy Panda Day everybody!

We have to see some kind of change soon. I have never seen the SERP in my niche be stuck in such cement like they have been this last 3 weeks, since I have been monitoring everyone's position.