1

I have a website that contains a lot of short (150 word) descriptions of key concepts. The concepts are organised hierarchically so a description of Cheese will be a subsection under the heading "Milk products" etc.

The problem is that the concepts are generally written so that they are best read as a page with whole sequence of concept descriptions. It does not make as much sense to read just one individual description (they don't quite stand on their own). However, the sequence changes depending on the user (an expert cheesemaker will want to read about detailed descriptions of Epoisse, while a non-expert just needs to read some basics on cheese making).

  1. What is the best way to choose the canonical pages for this?

  2. If it is to put each individual description as the canonical page, how I can make subsections of the pages showing a sequence of descriptions have a canonical link to the pages showing just one description? As far as I know the canonical properterty works only for page header.

  3. And is having each individual description as the canonical page really the best setup given that users will generally prefer sequences of descriptions?

Improve this question

3 Answers 3

Reset to default
2

You're not explicit about this, but I'm assuming you're asking specifically about using the canonical link element (CLE).

If so it's worth noting that Google (for one) don't guarantee to honour it; they describe it as a "hint and not an absolute directive". Because the CLE is intended to deal with duplication (and near duplication) rather than solve architectural problems, I'd expect there to be a greater chance of it being ignored when used in the latter way.

In other words it doesn't sound like you're solving a duplication problem, but a site or information architecture problem. So I'm not sure that CLE is the right tool for the job nor a reliable one.

In my opinion, it'd be much better to structure your site in such a way as to design the problem out.

Improve this answer
2
  • Your point about architectural problems is well taken. Any suggestions for where to start looking for this kind of issue?
    – tsk
    Mar 11, 2013 at 14:04
  • @tsk You could probably pose at least some questions here (it's outside of scope of current question), and perhaps on the UX branch of StackExchange too (they have an "information architecture" tag). If you're at a point where you'd struggle to ask precise enough questions, I suggest looking at existing sites for ideas. Failing that, web design forums and Google+ groups might be a good fit for canvassing ideas on alternate approaches.
    – GDVS
    Mar 11, 2013 at 16:22
1

A canonical is set per page and functions much like a redirect, whereas no redirection is taking place.

From your description I assume the following scenario:

  • Page 1 - Details about Spray-On-Cheese
  • Page 2 - Details about Goat Cheese
  • Page 3 - Details about Cow Cheese
  • Page 4 - Details about Goat Cheese Manufacturing
  • Page 5 - General Goat Cheese Info (exact duplicates of Page 2 and Page 4 on one page)

Google Webmaster Tools Help says this:

To gain more control over how your URLs appear in search results, and to consolidate properties, such as link popularity, we recommend that you pick a canonical (preferred) URL as the preferred version of the page.

Basically this means, you need to decide on which page should gain from backlinks, etc. In your case, I'd canonical all non-overview-pages to the overview-page. In your case, that would be:

  • Page 2 -> Canonicals to Page 5
  • Page 4 -> Canonicals to Page 5
  • etc.
Improve this answer
6
  • Thanks - but the trouble I have is that page 5 in your example is adapted to user preferences (normal users might only see the details on goat cheese while cheese experts see details about goat cheese and goat cheese manufacturing and the geographical distribution of goat cheese). So it does not make much sense to make it canonical either.
    – tsk
    Mar 11, 2013 at 13:45
  • the only case you need to consider is what Google crawls. User preference or not, what is the default for Google? That's where you need to canonical to.
    – David K.
    Mar 11, 2013 at 14:20
  • @DKOATED But surely whatever is the canonical will (assuming the CLE is obeyed) be returned in SERPs, so user preference is important.
    – GDVS
    Mar 11, 2013 at 16:27
  • @GDav I don't understand. The canonical tag with it's attribute is merely an indication for Google that some duplicate content is present and that the canonical URL should get the attribution (and no penalty for the current crawled page). Since Google does not set any user preferences, nor does it login to a page, a logic based on user behavior is nonsense. Perhaps I misunderstood the question though or the term "user preference" is unclear in the question...
    – David K.
    Mar 11, 2013 at 21:03
  • @DKOATED My point is simply that Google will prioritise the content identified as canonical in SERPs. As a result, logic based on user preference (i.e., what a visitor with specific interests should see) is anything but "nonsense", if we assume that several combinations and permutations of the same "building blocks" of content are functionally discrete. Hence my suggestion above that it's probably more of an architectural issue.
    – GDVS
    Mar 12, 2013 at 11:43
1

All duplicate content on the web is not bad and there is no need for a zero tolerance policy. It is fine to repeat content on your site when it is in a context that is appropriate for users. Usually, Google will choose to index a page that contains unique content even it contains some duplicate content. If Google finds a page that has nothing but snippets that appear on other pages, then I may choose not to index that page at all. Having all content on your site appear on two or three other pages is not going to get your site penalized or cause Googlebot problems when crawling your site. At worst Google will just choose not to index some of your pages.

The canonical tag is most useful when there are a large number of possible variations in URL that all return the same content. Some examples are:

  • Each user gets a tracking parameter such as a session id like /index.html?ss=t3te9893t3e
  • There are paramters on the the url that control how items are sorted on the page like /products.html?sortby1=name&sortby2=price&sortby3=rating

In those cases, the content can be duplicated hundreds or thousands of times and the canonical tag keeps Googlebot focused on a single version of the page.

Improve this answer

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.