How to avoid duplicate content on internal search results page?

[ShaqD]

Hi,

according to Webmaster Tools and Siteliner our website have an above-average amount of duplicate content.

Most of the pages are the search results pages, where it finds only one result. The only difference in this case are the TDK, H1 and the breadcrumbs. The rest of the layout is pretty static and similar.

Here is an example for two pages with "duplicate content":

https://soundbetter.com/search/Globo

https://soundbetter.com/search/Volvo

Edit: These are legitimate results that happen to have the same result.  In this case we want users to be able to find the audio engineers by 'credits' (musicians they've worked with).  Tags.  We want users to rank for people searching for 'engineers who worked with'.  And searching for two different artists (credit tags) returns this one service provider, with different urls (the tag being the search parameter) hence the duplicate content.

I guess every e-commerce/directory website faces this kind of issue.

What is the best practice to avoid duplicate content on search results page?

[LesleyPaone]

No-index your search results. Most platforms do it by default to eliminate that error.

[seowoody]

Hey Prestashop

To add a little more clarity - would you:

a.) add /search/ to robots.txt, like so:
Disallow: /search/

or

b.) add noindex/nofollow at page level: like so:
in the search results page template.

I would opt for option b, but it would be interested to hear your thoughts too and why.

Thanks,

[LesleyPaone]

I would add it to the robots.txt file. Depending on how your cms is set up, you can grab the search string from the current url and also use the presence of it to fire a no index as well. I wouldn't do a no follow however, there is nothing bad about following it, it is just the indexing of the search pages.

[ShaqD]

Thanks Prestashop for your answer.

Is there another solution other than no-indexing all our search results?

Like many sites (yelp, tripadvisor and others) our search results help drive traffic.  They aggregate the answer to questions that are asked in searches, such as 'recording studios in london'.

https://soundbetter.com/search/Recording Studio - Engineer/London, UK

[seowoody]

I wouldn't do a no follow however

I agree. My solution was to use NOINDEX, FOLLOW.

[LesleyPaone]

Sorry, I misread it. I think either or in regards to the robots or on page is applicable. I think the on page would make them fall out faster though.

[LesleyPaone]

It really depends on your developers and your budget. I do development and SEO, so this is how I would handle it. On searches that are returning just one result, I would put something in place to see how many results are returned, if it is only one result returned, in the head of the page I would set the canonical url for the search page to the actual page that is being returned as the result.

If more result is being returned, you can handle that in many different ways. One way would be to create a pseudo category out of the results page. I would use this sparingly and only for popular search terms. But you could have an extension written for your site that can give you some on page control of the text, the url, the meta areas, and things like that. I wrote a module for a platform I use a couple of years ago that does something like it. http://blog.dh42.com/search-pages-landing-pages/ You can get the gist of the idea by reading about it there, but that is one good way to handle a limited number of them to get them to rank better.  I would not do it with every search result though, you might get a penalty.