pagination for product listing. what to use? canonical or rel-prev-next or do nothing?

Question

I want to make sure my product listing is 10 products per page which are not in a series.

http://support.google.com/webmasters/bin/answer.py?hl=en&answer=1663744&topic=2371375&ctx=topic

They have explained how to use canonical or rel prev for pagination when a long page has been divided into multiple page and the multiple pages becomes a series were as my condition is not that. they are unique listing which are not related to each listing... all the listing links leads to a product profile page.

So lets say my site is all about cars and i have a "Used Audi" page with 1000 Audi's for sale. There are 10 used audi cars on each page so there's 100 pages in the series. If i start to utilise Rel="prev" and rel="next" should i set page 2 onwards as "index,follow" or "noindex,follow"? The content on Page 2 all the way to 100 only changes ever so slightly as different cars will be for sale on different pages but from a "Panda" point of view the pages are incredibly similar as they'd hold the same meta data as page 1 in the series along with duplicate reviews & news etc.

I want Page 1 in the series as the "Main" page for Google to send users too and i don't see the point in Google indexing page 2 > 100. What's everyone's view on this?

Lastly with the rel="canonical" tag should page 2 to 100 all point back to page 1 in the series or the individual page itself? E.G: /used-audi/page-3/

Feedback would be massively appreciated.

Answer 1

The number 1 question: Do you want your pages (products) to be found by the search engines? Do you want the bot re-visit very often and timely to find all your latest updates and provide it to searchers? Then act on visibility, not on artifical limits and noindex.

If the Audi example is similar to your actual page you have a category each as a starter and you can make use of rel=prev/next. It's a hint to Google that there is a relationship between the items (Audi-cars). If you don't do it, Google uses it's own heuristics. But if you have this tool at hand, why rely on the algorithm? Give it a try.

Meta-Data for every page: If you slightly change title and description, which should be very simple (e.g. "Audi cars - page x"), they are unique enough, based on the product listing which is unique anyway. I assume the other content (teasers, banners, etc.) that stays the same is in terms of quantity in an acceptable manner.

Add semantical hints to your product entries like date and time and make use of relevant markup to further optimize the listing.

Concerning the "canonical": I personally would not use it unless there's a strong need. If done wrong it can do more harm than good. As the documentation (your link above) says, it's an orthogonal concept to rel=prev/next.