Did Google's Farmer Update Positively/Negatively Affect Your Search Traffic?

[ilyaelbert]

no, and I noticed that merchantcircle, one of their 'top 25 biggest losers' still ranks quite well for long tail service-industry queries.

[Matthew_Edgar]

I've not seen any change either. One of my clients has a news site that is pretty ad heavy. While the site does run some non-original content (wire stories for example) at least 60% is original content. Overall Google traffic for that site is up 4% in the last week - with 10% more keywords delivering traffic than the week before. That is right on average with our normal week over week increase.

[ilyaelbert]

actualy I lost about 10 places on my main keyword; strangely NO other keywords were affected and long tail traffic did not get adjusted as well.

[DanDeceuster]

I haven't seen any impact on my sites or on my day to day searches. I think most people took the line that 12% of queries were affected to mean that 12% of websites were affected. Far from true. I think while 12% of search queries may have seen the rankings visibly change, this does not mean 12% of Google's indexed sites were affected in some way. I think this was a handful of sites that accounted for a lot of long tail search results.

[FrankWickers]

On many of our client sites we've actually seen positive impacts from the farm updates.

Previously, some of our clients were being beat out for the top few positions by some content farms for various long tail search terms.

After the Farm Update we've seen those content farm page results drop off into oblivion and our client sites have stepped right up into their positions. Our client's have gained dozens of new 1st, 2nd, and 3rd position SERP results on long tail keywords.

So in summary - none of the websites we manage/seo have been hurt by the Farmer Update. We have, instead, been rewarded from it.

[daveschappell]

We (www.TeachStreet.com) have been pretty negatively hit, with a reduction of ~44% week over week (comparing Thurs-Mon vs a comparable prior period).  We're trying to be calm, and find out what's driving it -- we think it's because we are a directory of classes/courses, and many of these classes can be found on the sites owned by our customers... but they're all formal relationships (not scraped content, etc.) so we're not sure what to do.

Any ideas welcome / appreciated.

Cx4QP.png

[mikecp]

Dave,

There is a considerable amount of research being done by Moz and the Moz bloggers at the moment on this topic. A blog post is planned once the team reaches a conclusion. I would expect it within the next few days (if not tomorrow).

If the post doesn't answer your question, you might think about posting your own separate question to get feedback from the community.

Mike

[randfish]

Teachstreet doesn't fit the profile of a lot of other sites that got hit. Were some pages affected and others not? If you could show off a dozen or so of affected vs. not, that could really help sort out the issue (and possibly give us a roadmap to help).

That sucks Dave! Teachstreet has been getting so good lately, too.

[StoreCrowd]

We got hit at StoreCrowd by around ~40%, whilst it's not devastating for our business it does raise some concerns about what we're doing (in comparison to competitors).

I've made a few observations:

1. Many of the websites that I've seen hit appear to have a large number of indexed pages (TeachStreet you have over 1M+ for example)
2. It appears to have hit sites that don't have a high amount of unique content to page ratio. For example we currently have a lot of blank "placeholder pages" & we also split out our merchants into 4 areas (store, coupons, deals, reviews).
3. The penalty appears to be sitewide, we have a large number of links & quality on our blog but even these pages have been hit hard.
4. The dropoffs in rankings can be a few spots or ~30 spots, I can't explain why this is - for less competitive keywords the dropoff appears to be less. This leads me to think this ain't a penalty but Google is simply reranking based on new factors.
5. I don't think this has anything to do with links.

What we've done so far:

1. We're working on increasing the unique content to page ratio - we've noindex,nofollowed all pages that have 0 content (or the placeholder pages I spoke about)
2. The next step is to further increase the uniqueness of our tag pages & store level pages.

But, we have competitors that have a lot more duplicate content than we do & they seem to be fine. So, we're merely testing & speculating at the mo!

Happy to hear any suggestions.

[StoreCrowd]

Just to follow-on from my last post, apparrently Google is working on a fix 

[RealSelf]

Our Google.com traffic dropped 29% last week, and we get a fair amount of International traffic too (24%), so we'll probably drop further when the change is replicated to .ca, .co.uk, etc.

Looking at our keyword traffic (or our SEOMoz ranking report), it's really clear that our biggest decreases in visits came from stem keyword searches like [botox] or [invisalign] where we were on the first page of results.  The ironic thing is that those pages have low bounce (17% and 27%) and no AdSense ads. We removed the AdSense ads from the landing pages a few months ago to simplify the initial visitor's experience.

It looks like Google is treating those pages as a list of blogs instead of a TripAdvisor hotel listing.  Many of the sites that were hit in this update were faux blogs targeted to high CPC keywords.

Since last May, we've been frustrated that Google keeps ignoring our meta description in favor of the first review snippet on the page.  (Bing doesn't do this.)  Even if our rankings are down, we at least want to present an accurate view of the thousands of reviews, photos, and answers that our site has and not be treated as one person's blog.

We've tried a lot of thigns to get Google to pick up a better meta description, but at best it only works on our site search for a few hours and then Google takes user text from a freshly updated review.  At this point it looks like our only option is to pass the first words of each review in a JSON array and display it using Javascript, possibly on mouseOver.  Its frustrating that we have to do such contortions, but as Fred Wilson says, it "sucks being a Google bitch."

organicfarm.png

[daveschappell]

Thanks for your response Stuart (and Tom/RealSelf and others as well) -- some extra color from us:

• Yeah, we have a lot of indexed pages.  However, there isn't much we can do about it, as we truly do have more than 500,000 class listings alone (not to mention Teacher and School Profiles, category pages, etc), and these are all unique in some way (price, date, geo-location, etc.).  You could argue that we could include that all on one summary page, but then we'd equally frustrate users who are looking for their exact match.  We decided to focus on humans in this instance, vs. the needs of bots.
• We're working to reduce some placeholder-like pages.  For instance, we've been creating pages for something such as 'Wichita, KS Programming', but it may only have Online Classes.  In the next 24 hours, those pages (that don't have any local/in-person classes) will redirect to the online/non-geo versions of Programming pages.  Here's an example of one of those pages:

http://www.teachstreet.com/wichita-ks/sewing-fabric-arts/50564-385

After our change, this will redirect to this 'online class' page:

http://www.teachstreet.com/sewing-fabric-arts/classes/385

• We've also seen the impact to be pretty much sitewide.  And we can't identify any specific geographies, categories, or page types, that have been specifically impacted.
• As part of our review, we HAVE found some sites that looks to be creating some pretty eggregious copies of our data (for instance, the family of sites owned by www.hellometro.com, that spawns 1,000s of similar sites like www.helloseattle.com, have our content on them, with no link-backs).  So, we submitted those types of sites to Google for review.
• We also resubmitted TeachStreet to Google for consideration, in Webmaster Console.
• We're removing some legacy 'seo spammy-type content' that we've had on the site since we launched, that we've never bothered to remove (meta-keywords, top-of-page-category descriptors, some excess footer links)
• We had removed some 'Article' and 'Q&A' type content from our Category/Subject pages (to increase their page-load speed)... we'll be moving some of that back, because the content is unique, and high-quality, and also because we think we can do so, without impacting page-load times

Any other ideas?

Dave

[daveschappell]

I look forward to seeing the Moz Reply -- I've been refreshing your blog since yesterday!

[RealSelf]

Good ideas, Dave.  Your housecleaning ideas prompted us to delete 2,500 old skin care product pages that we had converted to static pages after switching off Drupal.  It constituted 1% of our pages and visitors, but the poor metrics from those pages could have acted as a bad signal (2,500 pages is a lot for a human to create) and tarnished our overall site.  1% rotten is still too much.

Good idea about bringing back more content onto the Category/Subject pages.  We're going to do the same for our main landing pages.  The data I've seen indicates site speed seems is only used to break ties between pages with a similar ranking.

• Eric (CTO at RealSelf.com)

[RealSelf]

Can someone who didn't see a drop confirm or deny if the "Requesting reconsideration of your site" link is in the yellow Help box on the left side of your Webmaster Tools?  That link just appeared for us, and I wonder if that's because Google has set some site-wide penalty on our site?  It seems more likely that Google added it for all users after the update, but it would be significant if someone doesn't have that link in Webmaster Tools_._  We don't want to further anger the Google Gods so we're going to wait for the updated algorithim instead of requesting reconsideration at this time.

reconsideration.png

[Blargh123]

Looking at the traffic, our sites have seen an increase, seems to be assisting getting rid of some of these affiliate farmers for ecom as well.

[kurus]

Has anyone seen a recovery after taking remedial action?

[Blargh123]

Prob wont be an imediate recovery, there will certianly be a peroid of reduced flow.  but legitimate sites that are not using "farmer" tactics will likely see signs of recovery soon when google indexes enough of the new links.  Set up 301's on all your old links point them all to related pages in your new structure..... it could be done just rattleing off ideas here.

[Francisco_Meza]

Woot Woot! I don't see eHow above me anymore!

[sichristie]

Hey Dave,

I would love to know what kind of errors and warnings were being thrown up in crawl diagnostics on you seomoz campaign.

I am responsible for 3 large E-Commerce sites that have all the normal duplication issues, too many links etc. We rank fairly well, but the algorithm change hasn't hit our shores yet. So I'm curious what errors and warnings you have, so that I can raise the argument for acclerating our seo changes in-house.

Would be interesting for us all to know so that we can attribute a possible correlation between these and the Farmer changes