Welcome to the Q&A Forum

Hello Jim,

Ben's suggestion is typically what I would suggest as well, but I understand you have a situation that is unique to companies that sell parts to items like boats, vehicles, appliances, etcetera... in which case a single product page with variables won't work because you're dealing with completely different skus / part numbers.

In this case I like your idea of creating a single landing page and then using the rel canonical tag on the others. This way the other pages are still available to users who need a specific product part number, but search engines can see that these are all URL-variations of the same content.

Another, similar option would be to use variables in the URL and instruct Google and Bing to ignore those variables and treat them as if they were the base URL. Example:

www.savvyboater.com/store/p/Cover-for-V-Hull-Fishing-Boat-with-Side-Console-O-B-17-X-92-.aspx**?variant=72216P**

www.savvyboater.com/store/p/Cover-for-V-Hull-Fishing-Boat-with-Side-Console-O-B-17-X-92-.aspx**?variant=72217P**

And this would be the canonical / base URL:

www.savvyboater.com/store/p/Cover-for-V-Hull-Fishing-Boat-with-Side-Console-O-B-17-X-92-.aspx

That may be a problem from the development side because your standard URL includes the product's unique identifier at the beginning, as in:

http://www.savvyboater.com/store/p/2100-Cover-for-V-Hull-Fishing-Boat-with-Side-Console-O-B-14-X-74-.aspx

An example of a problem that could arise is if you decide to just choose one product to be the main one, as in:

http://www.savvyboater.com/store/p/2100-Cover-for-V-Hull-Fishing-Boat-with-Side-Console-O-B-14-X-74-.aspx**?variant=72217P**

http://www.savvyboater.com/store/p/2100-Cover-for-V-Hull-Fishing-Boat-with-Side-Console-O-B-14-X-74-.aspx**?variant=**72216P

But then what happens if 2100 goes out of stock or is removed? Would every other model then be inaccessible at that URL?

I'd have to know more about this before helping with that issue, but it may not be a problem at all. You'll have to discuss options with your developers and go from there.

Good luck!

Hello Peter,

Best practice is to avoid looking like a thin affiliate site. That means making the content very useful and unique, obviously, but also linking out via non-affiliate links to other resources as well. In situations where you have several followable links going out to high-quality, trustworthy sites about the topic I have not found it necessary to hide or redirect affiliate links. I used to advise putting the js that controlled the redirect into an external .js file within a /script folder and then blocking that folder, but have found that Google still executes the javascript (they have to or they wouldn't be able to navigate half the sites out there) even if the file is blocked.

However, if your site is mostly affiliate links (like a review site) and you don't feel comfortable with, or don't have time, to link out to trusted resources and create unique, useful content that provides a lot of value, then I would recommend going ahead and using some type of redirect script. In my opinion you'd be better off going with an ad management tool of some sort over an known affiliate redirect script or plugin, which may raise a red flag right away.

Long story short: There is nothing wrong with sending links through a redirect first if you want to track the clicks and other metrics. I just wouldn't use a tool called "affiliate cloaking script" or something like that.

Good luck.

Nathan please see my response to this question in the Private Q&A section. In short, I don't think this has to do with Google's goof-up a couple weeks ago, or with some of the other shifts, but I do think you are in violation of some of their guidelines on a pretty massive scale, although unintentionally and with good reason. I'll go ahead and paste my answer below, but the Private version includes links and other information that could identify the site.

Hello Nathan,

There were some pretty major shifts on the 16th, 19th and 24th-26th so the time-frame raises a red flag. However, with the little information I have (basically your indexation count) I think you may have a problem that isn't related to any of the recent algorithm shifts.

Assuming you are talking about ####.com, I show 462,000 indexed URLs from that domain. Nearly half a million URLs is quite a bit to index and you can imagine that Google might want to thin that out a bit and focus only on the ones that are important and original.

If you have 460k + pages, but most of them are duplicate content or significantly duplicate content, and most of them have no external links, this would put you in danger of being affected by several different algorithms/filters/penalties put in place by Google to keep such pages from bloating their index and outranking what they think to be "better" content.

That can be a hard pill to swallow because you know your content is good and people like what they find there. But let's look at this from the perspective of an impartial machine...

The following EXACT phrase appears, word-for-word, on about 22,800 different pages, most of them from within your site: "PHRASED REMOVED FOR PUBLIC VIEWING"

The following is typical of the "Related Terminology" section of your pages, which could be interpreted by Google as being keyword spam: "You may have searched any one of these terms to find this product: Keyword1, keyword2, keyword3, keyword4, keyword5, keyword6, keyword7, keyword8, so forth and so-on for a few dozen keywords".

A lot of these pages, possibly most of them, have very little unique/exclusive content. Instead, they list out features and uses from a database. Because of this you have many thousands of pages all with the same potential pool of words, each choosing to show more or less the same words in various orders and combinations. Furthermore, it is obvious that the pages are generated by a machine. Looking at this example, a "better" page would be one that has an introduction telling the visitor what "Polybenzimidazole (PBI) is and what it's used for in paragraph form, in addition to the list of features and uses: http://www.###.com/###/ . I'm sure most of your users will know what PBI is for if they search the site for it, but remember we're dealing with machine algorithms designed to detect spamming attempts, such as article spinning, which uses pretty much the same technique of switching around the order of words to generate hundreds or thousands of "articles" from a single original.

I wouldn't venture to provide specific advice on how to fix these issues without knowing more about your business. My suggestion is to look for a reputable outside SEO agency who can help you overcome these issues, which may involve removing a lot of pages from the index, allowing more content to be seen on each datasheet, or some other measures.

Good luck!

I can see right away that not everyone agrees on how to optimize a category page. This is just my opinion based on experience, but I really don't think that 500 words, or several paragraphs, is necessary or even desirable on a category page. And I don't recommend putting the copy down at the bottom either.

If you use the word "SEO copy" to explain a chunk of text, right away that should be a red flag to you. If you claim that copy is good for users, why would you put it way down at the bottom? Because it kills your conversion rate? Well then it's not good for users is it? What's good for users will improve your conversion rate. Remember that.

In my experience a category page only needs about two-three useful sentences with appropriate keywords to rank. And by "useful" I don't mean just saying "This is our Green Widgets page where you'll find the bets green widgets in the world. We have big green widgets, small green widgets and cheap green widgets."

I mean something like "Find the best green widget for your needs by using our filtering options above. Choose to see green widgets sorted by price, color and popularity, or simply browse the offerings below. Call 1-800-Green-Widgets or click the 'Chat' button to the right if you have any questions."

If it is a complicated topic, like something scientific or technical, you may consider adding more copy for the users to help them choose the right brand / product. In this case, a drop-down "Read More" type of div works well, as does a link to a larger "guide" on another page. If they don't know what they want yet, maybe your category page is too far down in the funnel, in which case sending them to a separate "guide" could be beneficial.

I'll leave this topic open for discussion for awhile since others my find some very good reasons to disagree. In the end, what you really need to do is test out a few different options on different category pages and go with whatever works best for your users and your site.

Hello SKE,

I've dealt with this or similar situations several times and have found the following to be the best course of action:

1. Report the scraper site to Google using the DMCA complaint form:
https://www.google.com/webmasters/tools/dmca-notice?pli=1&

2. Take measures to protect yourself against future scrapers.
http://www.affiliatebeginnersguide.com/articles/block_bots.html
http://www.seomoz.org/blog/postpanda-your-original-content-is-being-outranked-by-scrapers-amp-partners

Good luck!

It seems like maybe we're getting off topic. Why does the affiliate have to suffer in order for the merchant to succeed and vice versa? The real problem here seems to be that you are giving your affiliates the same content you use on your own site. Either make them write their own content or change what's on your site and feed them the old content. It is more work, but you could start slowly by writing fresh (exclusive) content on your site for the most important products. This would give you the ability to test it out unless there is some site-wide (e.g. Panda) issue going on .

As both an affiliate and a merchant, I've always found it best if each has their own content. For one thing, the affiliate site sits earlier in the funnel so it would make sense that they wouldn't be using the same message as the merchant sites product detail page, which is about as far into the funnel as you can get without being inside a shopping cart.

If you are unwilling to do this I think EGOL said it best:

"However, if your rankings are falling it could be competitors (and your good affiliates) are working harder than you."

If you really want to be seen as the authoritative version when there are multiple sites with the same content, the biggest factor in my experience is simply links. Domain authority plays a role too, but a couple of deep links into your product page will make all the difference. I presented about some ways to get links into product and category pages at SMX West, 2012 (link to presentation) and also wrote a blog post about it here on SEOMoz. Hopefully that will help you get started, but there's no easy, scaleable way to do this that isn't a little bit on the gray side. To do it "right" in Google's eyes just takes a lot of elbow grease.

One last thing. As an affilaite there is no way I would agree to putting a cross domain rel canonical or rel author tag on my site that points to the merchant's site. You would lose any affiliate worth their weight in salt that way.

To sum things up for Gavin, here are your two options as I see it, but they aren't mutually exclusive:
1. Rewrite your descriptions and either give the old descriptions to affiliates (e.g. have a database with two different descriptions for every product) or stop giving descriptions to affiliates and make them write their own.

2. Build more external links into your product pages.

Hello James,

I know this sounds cliche, but do whatever is best for your users.

While I don't agree with the idea that "almost identical content" is ok for thousands of pages on an eCommerce website, I do agree with the statement that you shouldn't make things more complicated than they need to be.

As a general rule, i try to make one product page per sku. If a single sku has multiple variants, I tend to let those variants be selected on the product page. A simple example would be a pair of pants that could be purchased in small, medium or large, as well as in black or white, for a total of 6 possible combinations - but all the same sku.

I like to keep the category information OUT of the product page URls because that creates all sorts of potential problems (like when a single product exists in multiple categories, or gets moved from one to another) and I like to put them all in a /product/ folder so I can easily diagnose issues by, for instance, searching for something like site:yourdomain.com inurl:product to see how many of my products are indexed quickly.

So the URL might look more like:

www.domain.com/product/brako-super1
www.domain.com/product/brako-super2
www.domain.com/product/brako-super3
www.domain.com/product/brako-super4

Once the user gets to that page they can then select options, like which model car they have. However, this may not be the right solution if each vehicle model has its own sku for that product. I'm assuming in this example that there are really only four skus, each with many possible variants.

Yes, this limits your ability to target long-tail, vehicle-specific searches for that product, but you may be able to rank for them anyway if competition isn't extreme for them and you list out on the page which vehicle models/years each product works with. You can put this in a tabbed description area or under a collapsable "See Vehicle Models" div so it doesn't muddy up the rest of the page visually.

In terms of category structure, you can set that up any way you like without affecting the product URLs if done this way. It sounds like your users my enjoy some type of faceted navigation or faceted search options. These are great usability features, but come with their own set of SEO issues that you'll need to look into.

Please let me know if I have missunderstood anything.

Good luck!

Smallbox,

This reply assumes that the "auto scroll feature" you mention is what is also known as a form of "progressive loading", though progressive loading can also be used to load unseen files like scripts and tracking pixels. On-scroll progressive loading of page content is a standard that more and more websites are using, including Twitter, and Google+ and so web users are getting very used to it. Most don't even notice that it's happening. Thus, I don't think it's a problem.

However, if you're talking hundreds or thousands of products, as opposed to dozens, you may want to provide an alternate viewing option, such as pagination. In that case, you would probably use the View All Canonical method as discussed here, providing your "view all" pages load all of the above-the-fold content within about 3-4 seconds or less: http://googlewebmastercentral.blogspot.com/2011/09/view-all-in-search-results.html .

Here is a blog post and a presentation I did on building links into product pages. Hopefully there will be enough tips between these two and Rand's Whiteboard Friday post (Devin linked to below) to move you forward:

SMX West 2012 Presentation:
http://www.seoverflow.com/link-building/seoverflow-er-everett-sizemore-speaks-at-search-marketing-expo-smx/ 

SEOMoz Blog Post:
http://www.seomoz.org/blog/building-deep-links-into-ecommerce-pages

Hello Georgios,

Activitysuper has a good point about choosing some of the best ones as testimonials for conversion rate optimization.

However, there is a chance you could get the third-party review data supplier to noindex their version of the reviews/content. It depends on what your contract says. If I was paying a company to handle the reviews for my product, like a BazaarVoice for example, I'd expect to have exclusive rights to that content.

To be specific though, we'll need more information. Where will these reviews go? If it would be 5,000 new pages of duplicate content, that's an issue. If they would be put onto product pages there are many options. You could put them in an iframe so you show the content to users but avoid the duplicate content issue. You could show the best three or four on the product page and then show the rest on a separate page that you don't allow to be indexed.

You are asking a great question, but I think some more information is necessary in order for anyone to definitively answer it.

Hello Caroline,

This really depends on the scale of the problem. If we're talking thousands of product pages or more then that's an issue we'll have to discuss in more depth to see what options there are (noindexing some of them, making them the same sku and offering them as a variant so you can put all on one product page like clothing stores do with colors and sizes, etc...).

However, if you're talking about a few hundred or less my advice would be to take the time to get creative and write unique titles and descriptions for each variant that has its own page. It will be worth the effort in the long run. To Naghirniac's point, you could think of this as an opportunity to target more types of keywords.

Please let me know if I can further elaborate, in which case I'd need to know how many products we're talking about here.

Thanks!

Not all links are created equally. This includes many factors, but one of them would be links that are contained within the site template, such as global navigation elements, whether in the header, sidebar or footer. It has been my experience that a link in the navigation does send the message that the page is important, but may not necessarily trump an in-content link with good anchor text.

So to answer your question, no I wouldn't say the in-page links are "pointless". But I would say you can use some more in-content links (as opposed to relying on the catalog and navigation links to pass anchor text), and that your below-the-fold content seems very over-optimized to me.

Rather than listing out links like on this page, you might try to weave them into some content: http://www.coolpooltables.com/categories/Darts/

Also, I would focus less on this kind of stuff and more on getting some useful intro copy on all category pages. I see that you have it for most of the top level categories, but I'd expand that strategy to the next level, such as http://www.coolpooltables.com/categories/foosball/foosball-tables/shelti-foosball-tables.html or http://www.coolpooltables.com/brands/InStroke.html .

Good luck!

Hello Alan,

If they are only doing this for Google then it is indeed cloaking, regardless of their intent. That may or may not land them in hot water, but I think Google has provided plenty of other ways to handle this situation.

First and foremost, it would be best if those session IDs and certain other parameters were put into a cookie instead of the URL.

As you probably know, you can tell Google and Bing how to handle the different parameters with their free webmaster tools so it would seem to me like that would be the best approach if you cannot get rid of the parameters all-together.

You can also put a rel canonical tag in the header that references the version of the URL without the parameters.

Force redirecting Googlebot is not a good idea. If, for instance, you wanted to force redirect by IP - as is the case with many global sites that have geo-specific landing pages - then that would be fine as long as you aren't making any special exceptions for Googlebot.

Did this answer your question?

Good luck!

Everett

Hello Ocelot,

I am assuming you have a site that has affiliate links and you want to keep Google from crawling those affiliate links. If I am wrong, please let me know. Going forward with that assumption then...

That is one way to do it. So perhaps you first send all of those links through a redirect via a folder called /out/ or /links/ or whatever, and you have blocked that folder in the robots.txt file. Correct? If so, this is how many affiliate sites handle the situation.

I would not rely on rel nofollow alone, though I would use that in addition to the robots.txt block.

There are many other ways to handle this. For instance, you could make all affilaite links javascript links instead of href links. Then you could put the javascript into a folder called /js/ or something like that, and block that in the robots.txt file. This works less and less now that Google Preview Bot seems to be ignoring the disallow statement in those situations.

You could make it all the same URL with a unique identifyer of some sort that tells your database where to redirect the click. For example:

www.yoursite.com/outlink/mylink#123

or

www.yoursite.com/mylink?link-id=123

In which case you could then block /mylink in the robots.txt file and tell Google to ignore the link-ID parameter via Webmaster Tools.

As you can see, there is more than one way to skin this cat. The problem is always going to be doing it without looking like you're trying to "fool" Google - because they WILL catch up with any tactic like that eventually.

Good luck!

Everett

I would imagine it would affect sales. There's probably no way around that. If they do this make sure call tracking is implemented so they can see what it does to their conversion rates.

As for rankings, Nakul Goyal gave the best answer so far with a first-hand account. He said they didn't change. But I would also imagine that you'd lose some sales from other channels that you can no longer participate in without pricing, such as Google Shopping, Amazon, and price-based comparison shopping engines.

There are other ways to block people from scraping your site for pricing besides requiring customers to call, so I wouldn't say that issue justifies this strategy on its own.

Good luck!

Everett

Hello Cherie,

You're probably already aware that it is possible to install WP in a subdirectory: http://codex.wordpress.org/Giving_WordPress_Its_Own_Directory . The problem, as Syed mentioned below, probably has something to do with Volusion being on ASP.net and Wordpress requiring a Linux/PHP host. In my experience, just about every decision you make going forward will be more difficult to get done if you go with an ASP.net site because a lot will revolve around having someone use Internet Server Application Programming Interface (ISAPI).

It is possible to set up the WP site on another host, but if you found Magento too complex, this might not be the best way to go either:

http://stackoverflow.com/questions/1848605/can-a-website-subdirectory-be-served-by-a-different-web-server

Can I ask why you want to move if your rankings good and business is going well? There are probably a lot of improvements you could make with the help of a good Wordpress developer.

If you really must move over to Volusion and want to keep the WP blog, this may be one of those times when the "Put your blog on your primary domain" rule of thumb is put aside for an exception.

Nwhite1,

To my knowledge there hasn't been any published study regarding the first-order effect of improved rankings after the inclusion of ratings on a product page.

My guess, and that's all it is, would be that the mere presence of them in-and-of itself would not improve rankings, but there would be several second-order effects, such as those you and the other responses have mentioned. Attempting to game the system by putting up empty reviews or fake reviews just to have something on the page could land you in hot water. Google has been working on the empty review page problem for awhile.

Local business page listings and Google Shopping results are totally different animals though, and the above statement does not apply there:
http://www.davidmihm.com/local-search-ranking-factors.shtml http://www.seomoz.org/blog/how-to-rank-well-in-google-products-search-a-big-list-of-places-to-get-reviews

I had half a rant on here about machine learning algorithms and how anything from the color of your background to the placement of your home button in the nav could theoretically affect rankings these days, but it seems like you may be after more factual, tested information, not more theories and ideas. The "fact" is I have never seen nor heard of a study that proves the mere existence of product reviews on a product page in-and-of themselves improves rankings, nor can I think of a way to test this without involving a myriad of second-order effects.

Hello John,

I have used several enterprise solutions like BazaarVoice, and solutions for open source content management systems like Drupal and Wordpress (there are plenty of those around, mostly free) but without knowing what platform you're on it's tough to come up with a recommended solution.

Since I haven't personally tested them out I don't want to list them specifically for fear that people will assume that is an endorsement, here is what I suggest: You may find what you're looking for by reading the following thread:
http://www.hotscripts.com/forums/script-requests/165-looking-good-review-management-script.html

And by looking into some of the solutions discussed on this page:
http://www.hotscripts.com/category/scripts/php/scripts-programs/reviews-ratings/

I hope that helps! And I'm pretty sure lots of people see this as a problem that requires a solution, thus I'm pretty sure there are some solutions available to you depending on what platform you are using.

Good luck!

Everett

It is unusual to endorse one of the shortest answers on the page, but Axel is to-the-point and, IMO, correct in this case. You don't want to paginate too heavily because that creates more clicks to get to all of your products. In fact, Google even recommends using a View All canonical page if it doesn't affect performance (load time) too much.

The first link anchor is what counts so I respectfully disagree with dittoeffect, unless you were to link to the image on the product page from the image on the category page using a Named Anchor hashtag (could be a good thing to test).

You don't want a bunch of iframes on your category page either. Keep it simple. You run an honest eCommerce site, not an uber-competitive affiliate website where you have to put links into a redirect script that goes through a directory that's blocked in the robots.txt file, etc...

Make the alt text and link text the same unless you are testing the named anchor link idea mentioned above.

And as Alan Gray said, test. These are all just opinions based on experience until you test.