Panda: Should I remove affiliate datafeed? Unrelated items?

[ianevans]

My site provides news and information on blue widgets. I attend blue widget events, publish blue widget stories, etc. That's what we are known for.

We were hit hard by Panda on July 11. 50%+ reduction in traffic.

For years now, I've been part of an affiliate program that sells blue widget-related items. They also sell green widgets and yellow...the whole darn rainbow. Since they had a datafeed, I found it easy to add over 1.8 million widgets.

A human visitor sees a site focused on blue widgets info. If they click through to the widget store that's where they see the rainbow for sale. But, of course, I realize now that Googlebot sees a site with 11,000 pages of info and 1.8 million pages of widget sales.

Will it help with Panda if I scale back dramatically and only use the info from the affiliate datafeed that pertains to blue widgets?

Yes, some months the sale of widgets has equaled or even exceeded my advertising. But what I'm on the web for, what I'm known for, my authority, is blue widgets. If Panda's slamming me for having a store in the back, do I need to close it or dramatically reduce it to regain visits to the info?

[Planet13]

...and 1.8 million pages of widget sales.

Question: Is the content on those 1.8 million pages of widget sales the same content that is appearing on other people's web sites? (i.e., is it duplicate content?)

[ianevans]

Sometime it's hard to talk in widget speak, but here goes... :-)

The datafeed is from one of the top sellers of widgets. They have complete downloadable feeds of all the widgets they sell. So, yes, the widget store on my site would include widgets other affiliates would have, and obviously all of us would have the same widgets the data provider sells on their widget store.

[Planet13]

So, yes, the widget store on my site would include widgets other affiliates would have, and obviously all of us would have the same widgets the data provider sells on their widget store.

well, I will leave it to larger minds than mine to advise you.

but if it were me, I would probably noindex those widget pages, to be honest. I just don't think have a HUGE number of duplicate content is going to be all that helpful.

If there is any way to create pages for the top selling widgets with your own unique content, that might be one way to go.

[johnhh]

Um maybe the debate should be - is Panda about "emphasis" if you have a red widget site - should you be showing blue widgets which detracts from the "emphasis" , or main subject matter.

No I don't have a conclusive answer as Pandarized pages I have re-empthasised have made no recovery.

Probably early days yet as machine learning takes many passes to improve - given the number of websites that exist on any topic.

[Sgt_Kickaxe]

50% reduction in traffic is too vague. Was traffic down equally across all pages? Was it down on a particular type of page? Did you have high traffic pages that accounted for most of your traffic and those were cut down?

I'd also like to know what your backlink profile looked like. If the sites that propped yours up took a hit it stands to reason you would too. Sometimes sites don't get enough exposure from a wide variety of sites and are susceptible to losing rank quickly.

My kneejerk reaction, without more information, would be to offer more. Find a way to make your pages more valuable and less duplicate.

[ianevans]

Well pages in the widget store represented about 62% of traffic at the end of June. After July 11, about 36%.

Most of our traffic is from searches on photos of the widget events we cover. 87% of our referrals come from Google.com and Google's country domains,

[Planet13]

We were hit hard by Panda on July 11. 50%+ reduction in traffic.

Well pages in the widget store represented about 62% of traffic at the end of June. After July 11, about 36%.

So the 50% drop in traffic is specifically a drop in the widget store, while the news and information site (which you sell advertising on) DIDN'T see a significant drop in traffic?

Or did the news and information part of your site see a significant drop in traffic, too?

Again, I will defer to others, but I don't think the problem would simply be that the news and information part of the site covers BLUE widgets, while the widget store sells GREEN and RED widgets.

I think it is more likely it would be due to the fact that your information comes from a data feed that is shared amongst numerous other vendors of widgets.

[ianevans]

Just looked at two main areas: info about blue widgets and original photos of blue widgets. If it was a stock chart, the 30 day average is moving sideways as the daily ups and downs do not appear to break a floor.

So yes, it's the extreme long tail of the widgets store that appears to have been hit.