Duplicate ecommerce product descriptions and ecommerce rankings (John Mueller)

[Robert Charlton]

Spinning the treatment of ecommerce product descriptions off from some questions regarding authorship that have arisen in our current discussion...

Google Authorship not (yet) a ranking factor, per John Mueller
http://www.webmasterworld.com/google/4613727.htm [webmasterworld.com]

There are still a lot of sites left out of Authorship (yo, ecommerce anyone? or just business or organization sites?) for it to be any kind of ranking factor.

There's no reason why authorship couldn't be a ranking factor for searches where authorship is relevant, and not for searches where it isn't.

In the John Mueller video Hangout that inspired the above, there are several inescapable suggestions that... at least with regard to product descriptions... authorship will not be relevant.

As I read it, Google is more or less officially bypassing the issue of who originated product descriptions, and will be looking to factors of site authority, query intent, and to other content on the ecommerce site to determine rankings...

Webmaster Central 2013-09-27
John Mueller
https://plus.google.com/events/cc0c9cadg9n7vrg4dp1c29k2gp0 [plus.google.com]

Click on the "play" arrow in the video screen in the upper left to get an Ajax video overlay. In the right column of that overlay, a list of discussed questions will appear. At the top of the list, at c20:53...

Q: Let's say a company uses product descriptions on their own site and at the same time provides the descriptions via a database to official retailers who might reuse them on their website. How can correct content attribution / ownership be ensured?

John Mueller: That's something where we probably wouldn't look into direct ownership of this content, but we'd rather try to figure out which of these URLs is the appropriate one to show for a user.

John then continues to examine possible areas of user intent... which include queries with local intent, queries with ecommerce intent, etc....

So, for example, if there's an online store that's selling a book, and it's selling it worldwide, and there's also a local bookstore that's selling the same book and on the site they have the same description as the big general online store, and if we can recognize that a user wants to find local content, then maybe we'll show them the local version.

And if we can recognize that the user doesn't want to find just local content but something maybe they can buy online, then maybe we'll show them the global version...

So it's not something where we'd say that if you wrote this product description your site will always be ranking for queries for that product description, but rather we'll try to show them the appropriate version that matches what we think the user is looking for. (c22:08)

Prompted by further questions, Mueller considers that the user might not just be looking for the product description, and that additional content beyond the product description, content like editorial reviews and user reviews, might be what the user is looking for, and that might make a difference in how Google ranks the site.

I also get the sense from questioning in the Hangout that Hummingbird's increased ability to determine query intent might be a factor here.

Note, btw, that John Mueller's answer technically covers only "syndicated" product descriptions, but I suspect the implications are wider. Mueller notes that matching the original author of a product description doesn't make as much sense as matching the author of a blog post or newspaper article.

He also clarifies that there's no "penalty" for having a dupe description, but he suggests that the ranking juice is likely to be elsewhere. He doesn't indicate whether having a completely unique description might provide a boost.

As I've discussed originality of product descriptions with clients who are manufacturers and distributors, and therefore theoretically originators, I've come to realize that there really is no way for them to determine what I'd call "essential" originality of their own marketing material. Many grabbed the online material from a brochure which might well have paraphrased other material about perhaps even competing products.

I'm guessing that tying rankings into query intent and other content on the site might not only be Google's most "relevant" way to satisfy the query... it might be the "only" way of handling the confusion of product descriptions that have built over the years.

[jamesMP]

Is it a fair assumption then, that the implication is that having duplicated boilerplate descriptions isn't necessarily a bad thing, so long as it is bolstered by value-added content such as reviews and editorial comments?

[ChrisWilson]

@JamesMP

I see evidence of this with the several eCommerce sites I manage. A lot of the sites will typically use the same, what we call, "romance text", provided by the manufacturer, but at the same time, we add a bunch of other details typically not found on competitors sites.

In more cases then not, we rank above most competitors when people search the specific product, like "Green Brand Widget"

[EditorialGuy]

Sounds reasonable--not just for boilerplate e-commerce text, but also for things like press releases that are often picked up and used verbatim.

As a user, I find it annoying to search on [topic] and find that many of the top 10 results are for the same press release on different sites.

[jamesMP]

That's what I was thinking... I tend to create unique product descriptions for all our widgets, but we're about to add a few thousand products to our relatively small eCommerce site and it simply isn't viable.

Creating a (unique) features/benefits list and using the manufacturer's boilerplate text may be the way to go.

[Awarn]

My issue with this mess is Google never looks to see if the site actually has the product for sale. They come out with do this do that but they don't utilize the data. With products it isn't that hard to tell. If the site has structured data they state right there for Google whether it is in stock. Most places also put N/A or Out of Stock also.

After reading this thread I looked at a competitor for just one category. They list 16 items with all the little write ups (they do well at that). My issue is they have 4 of the 16 items. Rank page 1 number 1. So I look at my site, same category, I list 10 items and have all 10 different products available and in stock (and the structured data that informs Google of that) yet I am 6 slots down. It just appears that Google gets baffled by BS. I get the comments all the time where the customer says you know you are the only one that has that. Oh I know it all too well anymore.

[EditorialGuy]

My issue with this mess is Google never looks to see if the site actually has the product for sale.

I think you have an unrealistic view of what a search engine is. Google Search isn't Consumer Reports. It isn't even Yelp. It crawls, indexes, and ranks content. Maybe Google Search will someday be capable of rating businesses on things like inventory, customer service, shipping times, whether they're using eco-friendly packaging, what their labor practices are, and so on, but for now, it's got enough of a challenge with traditional Web search.

[Awarn]

@EditorialGuy - Have you looked at Schema? From Google - The goal of a product rich snippet is to provide users with additional information about a specific product, such as the product’s price, availability (whether product is in stock), and reviewer(s) ratings and commentary.

However maybe your right an unrealistic view of this information engine even if all the data is there.

[netmeg]

Yea I'm not sure I really want Google setting themselves up as the ecommerce police more than they already are.

I think it's more than value added *content* that is required though. I think you have to have a unique business plan, or a unique perspective or a unique service. If a hundred thousand websites including Amazon are selling plain white paper widgets in boxes of 100, there's only so much useful or unique content that one can provide there.

So after you've combined all the products you can combine, and tarted up the descriptions as much as you can without them looking like they were run through Drunk Google Translator, you maybe have to get creative with other areas of your business model, to keep people coming back to your useful ecommerce site.