Are Article Pages Killing My eCommerce Site?

[Planet13]

Hi there, Everyone:

Can having a blog / article pages hurt an ecommerce site?

I know that people have often suggested having a blog on an ecommerece site to improve traffic and SEO to the ecommerce pages.

Let's say that we sell photographs of monuments around the world. The article pages would have information about those monuments (history, stats, etc.,)

The last two years it seemed to work for us. But we were hit with March 23rd, 2012 panda and I have a suspicion that the article pages are HURTING our site more than helping it, possibly because the user stats for the article pages are worse than the stats for the ecommerce section of the site:


User Metrics Compared when as a Landing page:

% of Total visits: (Articles) 31.73% / (eCommerce) 12.60%

Pages Per visit: (Articles) 1.70 / (eCommerce) 3.09

Average Visitor Duration: (Articles) 00:01:20 / (eCommerce) 00:01:19

% New visits: (Articles) 86.10% / (eCommerce) 85.11%

Bounce Rate: (Articles) 79.69% / (eCommerce) 51.96%

The thing that sticks out to me is that the bounce rate for the article pages is significantly higher than the bounce rate for eCommerce pages and that the pages per visit for article pages is lower than that of the ecommerce pages.

The other thing is that while the article pages to get linked to, which I am sure increases the page rank of the whole site, they don't drive much traffic or sales to the ecommerce setion of the site.

Any suggestions appreciated. Really need to get out from under Panda.

thanks in advance.

[Bewenched]

Google might look at these pages as "doorway" pages.

Now that would be a real shame.

on our ecomm site we do have some articles and technical information pages written to engage similar product owners that give our experience with products installed on certain widgets. I would think that the well written articles should show some authority on the particular products... since after all we are experts in that particular area.

[IanCP]

@ Sand

I have some sites with extremely high bounce rates (80%+), but sometimes that's just the nature of the content.

So have I.

It is often forgotten that if you have a page on repairing "Blue Widgets" and a visitor comes to the page and says: "Exactly what I'm looking for".

How long does it take to press Alt/File/Save OR Alt/File/Print?

I do it quite frequently myself. I can be in and out of a page in under two minutes. Bounce rate is not necessarily an accurate metric of site quality.

[Planet13]

@ claaarky:

thanks for your note about internal navigation.

I have been trying recently to make the navigation on the ecommerce pages more user friendly, mostly be reducing the number of links on the product pages so they only link back to the relevant category or to closely related products (sometimes to another closely related category).

Hopefully this will result in more page views per visit. It is still too early though to gain any real meaningful data from this yet.

[tedster]

reducing the number of links on the product pages so they only link back to the relevant category or to closely related products

This sounds like good thematic "siloing" to me - and it makes sense in a lot of ways as long as you have a good general top-level navigation for the site.

[bwnbwn]

Planet13 have you determined what key terms you lost. I am seeing were sites lets say had "wholesale wigets" and ranked for years with many other terms related to wholesale, ie wholesale brand name widgets, wholesale product name widget and so on. The update changed the site rankings from "wholesale" to "cheap". Now the same terms rank with cheap but not wholesale. Were before cheap wasn't a ranking term.

About every week the lost terms return for a period of time and then change back.

What I would do is determine what key terms I had and what key terms I lost. No way in this world would I delete good articles that is driving traffic.

I firmly believe the blog has nothing to do with the loss. I would look over anchor links you built up over the course of your optimizing the site. See if there is a relation to the links and the key terms you lost.

If you have traffic to the articles then you should seek some help on optimizing the articles to be sellers. This is good traffic you just need to optimize the pages to turn them in sales. Your blog should be an extension of the ecommerce site.

[viral]

A few people have mentioned navigation and one thing I found with quite a few of my clients is that they will add a blog to their site which has great inherent navigability but then don't maintain that same navigability in the eCommerce part of the site. If there is one thing that people who have a eCommerce site and are trying to position themselves as more of an authority in their niche do wrong it is this! There is a good Matt Cutts video (I can't find it at the moment)where he talks about navigation. He says (I am paraphrasing) sure Google bot can find your internal pages no problem but one thing Google is looking for is user experience. The biggest percentage of users tend to come in through the front page so how long does it take for them to find the info they are looking for from that front page. If it takes too many steps (corresponding to google bot taking too many steps)then that is no good. He goes on to mention how on his personal blog he has the archive widget in the right hand column specifically so that google bot doesn't have to take too many jumps to find content, even old content(and yes this apparently corresponds to users not having to click to many times to find content "although this is debatable"). Anyway i find that increasing interlinking between the blog and eCommerce areas actually drives improvements in the SERPS and that doesn't mean your standard drive the user to your sales page type linking. Here is the bit that makes almost all my clients cringe in horror I actually tell them to link from their squeeze page/sales page/landing page back out to relevant blog topics. Sounds crazy but it works. Panda is not only about content guys it is about how your whole website works. If you have bolted on a blog onto the side of your eCommerce site, then make it one site not 2 sites living on the same domain!

[Hope_Fowl]

I'd guess that the article visitors are searching for info on the topic, so they're mostly not shopping. But the articles seem like they have several things which should Google should like, so I'd tend to keep them. Actually, I'd be tempted to try "test" techniques, to try to find article page changes which convert visitors into product purchasers...as well as changes to merchandise which article visitors like (if you don't have T-shirts or mugs, add some and see if article visitors like those).

We don't know if Google has identified that there are two classes of pages on the site, much less if it is treating them differently. But I prefer to not underestimate Google, so the two kinds of pages might be getting different treatment somehow. But that shouldn't matter if buyers are still getting directed to your merchandise pages.

I will point out something which is probably unrelated -- Google Places just got shaken up; that probably did not affect this site, even if it really is about landmarks (and it probably is about something not related to locations). Google Places should only be relevant to the physical location of businesses, not location-relevant merchandise.

[Sgt_Kickaxe]

It's scary how damaging it is to the entire internet when Google goes bonkers isn't it? I mean, they make a change, one of 500 a year, and a formerly top ranked site is drawn under water and you're to learn how to swim on the spot, in the midst of panic, or drown? And even if you do learn, and swim well, you've been sinking the entire time and will need to wait to breathe again. It's ludicrous.

If you are happy with your content and your visitors enjoy your site who cares what Google thinks, let them replace you with spam if they see fit because it looks bad on them, not on you. This is one case in which I'd warn not to over-react. Do some minor A/B testing with improvements to a few pages and see what happens.

[zehrila]

Host your blog on a subdomain, use it to compliment your main index, by linking to product pages from your blog posts.

[Planet13]

@ tedster

This sounds like good thematic "siloing" to me - and it makes sense in a lot of ways as long as you have a good general top-level navigation for the site.

The navigation is more limited, I would say. Most of the category level pages don't link to the other major category pages. They would link to the home page and a few other closely related categories.

I have "flattened" out the navigation a little bit since being hit by Panda (and penguin), too, so I removed a few of the subcategories and merged them into their parent categories. I only had about 8 total subcategories, but I felt that it would be better to have larger categories that had more products per category and were less "optimized" for a key word.

An example might be where I had a blue widgets category and a porcelain blue widgets subcategory. I merged the two categories and was left with just the blue widgets category.

@ bwnbwn

Planet13 have you determined what key terms you lost.

On March 23rd it was a rankings drop of about 3 to 4 spots "across the board." So it was the main keyword "blue widgets" and long tail that had widgets in it. Plus, it was words that were unrelated to widgets (or blue) at all.

I would look over anchor links you built up over the course of your optimizing the site. See if there is a relation to the links and the key terms you lost.

Although possible, it is unlikely, but I will certainly look into it.

Anyway i find that increasing interlinking between the blog and eCommerce areas actually drives improvements in the SERPS and that doesn't mean your standard drive the user to your sales page type linking.

I should mention that, indeed, when appropriate, I do link from the ecom pages to the blog pages. I tend to agree with your premise; if it is good for the users, then it is good for SEO (which I think is what you are saying).

I'd guess that the article visitors are searching for info on the topic, so they're mostly not shopping. But the articles seem like they have several things which should Google should like...

Suppose that google is taking user behavior more into account. They might look at a site and say, "This site has authoritative articles, but people AREN'T following the links from the articles to the ecommerce pages. Maybe the amount of 'authority' credited to this site should be diminished?"

Maybe the amount of Page Rank that is flowed from one page to another is reduced if google sees minimal clicking of the link by users?

[viral]

I should mention that, indeed, when appropriate, I do link from the ecom pages to the blog pages. I tend to agree with your premise; if it is good for the users, then it is good for SEO (which I think is what you are saying).

That is basically what I am saying. Make it a good experience for your users and it should be a good experience for Googlebot. However that is not always guaranteed! If you want to see how Googlebot views your website then go to it using a text browser. Yes they do still exist! If you can get around the site using a text browser and you dont have to drill down to deep to get to all articles. Then Google should like your structure well enough.

[claaarky]

"Maybe the amount of Page Rank that is flowed from one page to another is reduced if google sees minimal clicking of the link by users?"

Bingo!

That was my earlier point, @ viral's and a few other posters. Google is more interested now in what PEOPLE do on your site. If nobody ever (or very rarely) visits a page on your site, it's either not of interest (poor or out of date content) or people aren't finding it (bad navigation).

Check your Google Analytics. If you are already linking FROM your product pages TO the blog (and vice versa) but people are still not ever visiting huge numbers of your pages, you might be better off without those pages. The fewer pages you have, the more chance people have of finding all your content and the user metrics / user experience improves.

However, don't make your categories too big as this could actually make matters worse (if you overwhelm people they just look at the stuff at the top and then give up).

I'm going through this process right now. Identifying pages that don't get seen and looking at why that is (not promoted enough or bad content/products). If the conclusion is people won't visit those pages no matter how prominently we feature them, they will be removed.

[claaarky]

"Maybe the amount of Page Rank that is flowed from one page to another is reduced if google sees minimal clicking of the link by users?"

Bingo!

That was my earlier point, @ viral's and a few other posters. Google is more interested now in what PEOPLE do on your site. If nobody ever (or very rarely) visits a page on your site, it's either not of interest (poor or out of date content) or people aren't finding it (bad navigation).

Check your Google Analytics. If you are already linking FROM your product pages TO the blog (and vice versa) but people are still not ever visiting huge numbers of your pages, you might be better off without those pages. The fewer pages you have, the more chance people have of finding all your content and the user metrics / user experience improves.

However, don't make your categories too big as this could actually make matters worse (if you overwhelm people they just look at the stuff at the top and then give up).

I'm going through this process right now. Identifying pages that don't get seen and looking at why that is (not promoted enough or bad content/products). If the conclusion is people won't visit those pages no matter how prominently we feature them, they will be removed.

[viral]

@claaarky it is not necessarily the case that rarely clicked pages give no value to your site and they definitely do not reduce the overall authority of your site. There are literally thousands and thousands of pages on on searchengineland.com that haven't been visited in years! Yet you won't see Danny removing them in a hurry. There is one basic rule when trying to build authority.. MORE CONTENT = MORE AUTHORITY. Of course it has to be good content!

Matt Cutts recently said they don't use bounce rate or any of that kind of analytical data to rank sites! I personally think he is not telling the whole truth and anyway it is hard for Google to get that kind of analytical data without people actually using analytics and there are some big sites ranking well that don't use Google analytics at all. Yes they can get some of this data through chrome users and google toolbar users but Matt says they don't. Can anyone else confirm if Google do use Analytical data? I would like to be shown I am wrong.

[claaarky]

@viral That's a good point about old pages and I was curious about how they fit into the quality landscape if they're no longer being seen by anyone.

Perhaps if those pages were once very popular the historical metrics give them a quality score that stays with them forever and, as you say, the authority builds.

I think our problem is similar to Planet13's in that we added a blog because our SEO's at the time advised us to but we never really liked the content so we kept it hidden away. Consequently it's never had the opportunity to build up a quality score. I think we either need to push it into the limelight to build up a score (although some is no longer relevant) or ditch it and add new content that is well promoted.

So the formula is:-
1) Add great new content and promote it well within your site so it builds up a good score,
2) Repeat 1 until authority status is achieved.

And do not:-
1) Add content blindly thinking Google will find it and that's all that matters.
2) Repeat 1 until Panda strikes!

[outland88]

On March 23rd it was a rankings drop of about 3 to 4 spots "across the board." So it was the main keyword "blue widgets" and long tail that had widgets in it. Plus, it was words that were unrelated to widgets (or blue) at all.

If you profits are that dependent on that small of a drop what else can I say without still being polite. I thought we were talking some major rankings problem here. Case closed.

[Planet13]

@ outland88

I appreciate all your input and courtesy so far in this thread.

I thought we were talking some major rankings problem here.

Being a small mom and pop ecommerce site, a drop of 3 or 4 places across the board IS a MAJOR rankings problem for us.

It is the difference between being able to pay rent or not... which seems pretty significant to us (never been a fan of homelessness myself, doubt my wife and six-year-old son would be enticed by the prospect, either).

_{Also, I assumed that whatever is learned from this discussion could be helpful to others who have mega sites that lost mega positions as well.}

[outland88]

Planet 13 my apologies. I know many are hurting and shouldn’t be. Google though is the true culprit. Your demeanor shows you are a class person and I truly hope it works out for you.

[MikeNoLastName]

I've noticed a pattern, overall, of the type of sites that Google loves.
It's not so much content or topic as structure. It's not really something you can define scientifically or logically, but you pretty much need to "get the feel" for it. Kind of like reading enough of a single literary author to be able to start to write a novel like them.
Search for your keywords, related keywords and even a few totally unrelated ones, and on the top 10 results in each (ignore the top few which are manually recognized authorities), study how the main pages, menus and second level pages are interlinked along with titles, descriptions and keyword stuffing and and you should quickly see a common pattern emerge. Then emulate it ASAP. This is apparently what G thinks ideally ALL pages should be like. It's a very narrow, optimal, peak, from whence either side of the preferred peak quickly drops off. It's a natural by-product of using an objective programmed algorithm to judge ALL sites rather than an observant. emotional person. Of course this model changes monthly.

[IanCP]

@ MikeNoLastName

While I don't necessarily disagree with your theory, one possible flaw is if it works, will it still continue to work next week, next month?

Google for webmasters, to me is a bit like a dog chasing its tail.

[idolw]

I think you should keep the info pages on the site. The fact they do not convert readers into buyers is not a problem. They've been created to be informational part of your site with a hope that the readers:
- will like the content and perhaps save it / share it
- remember your name as an expert in your field, or at least, as the place that has the right info easily accessible
- when they finally are ready to buy, they might remember your site.

If that happens often - cool. If not - no problem, the pages are there anyway, they're a chance to say "hello" to the users. I would not remove them.

In fact, that's what big brands do. Sponsor events, contests, competitions, create topical sites for different projects with a small logo of theirs. They want the eyeballs, maybe a possibility to add the users to their remarketing lists to have a chance to show them some ads in the future.
Your blog/info pages are the same thing - a chance to establish some little contact. If people do not buy, perhaps you can show them some remarketing ads or ask them to sign up to your newsletter / facebook page / twitter feed.
What I see in my extremely highly competitive niche is the fact that big guys introduce Customer LTV to their bids. They expect a single customer to buy not once but 2,3 or 5 times from them in a certain period. That way, they can afford higher bids for ads / other types of marketing and screw the small guy. Small business must follow in their tiny niche or die sooner or later.

[claaarky]

I'm working on a Panda theory connected to user metrics (which I'm waiting for the mods to unleash so we can see what people think and develop/eliminate the idea) and I've been checking it with a few people with different types of sites.

It's obvious but people react differently to different types of sites. On a good ecommerce site the metrics are all about keeping people on your site until they buy and leave via the order confirmation page, whereas on an 'answers' site it's about answering a specific question so the job is done as soon as the question is answered.

Quality ecommerce sites produce different reactions from people compared to rubbish ones. People have different expectations from an ecommerce site. Seeing articles may confuse them - they're expecting products.

I think with Panda, Google has a profiled 'good' ecommerce sites (based on user metrics of sites they know to be good) and judges how people react to your content compared to that profile. If your site produces user metrics very different to Google's profile (which will change as their seed group of sites change) that might indicate your ecommerce site is no good.

In short, I now think too many articles would be a bad thing for an ecommerce site.

I would also look at where people leave your site the most. You may have a lot of people leaving on a lot of pages, which might also be an indication of low quality pages (for an ecommerce site that is). I'm waiting for Webmasterworld to publish my post about the possible connection between exit rate and Panda, and waiting, and waiting.

[Robert Charlton]

I'm waiting for Webmasterworld to publish my post about the possible connection between exit rate and Panda...

Is Panda all about Exit Rate?
http://www.webmasterworld.com/google/4469114.htm [webmasterworld.com]

[tedster]

The opening question - Can having a blog / article pages hurt an ecommerce site?

Now a light turns on for me. I worked with one e-comm site where it clearly did, because most of the blog "articles" were written only to get rankings and not really to do anything for visitors. Think eHow content farm published on a blog. Rankings began to nose dive soon after.