Sitemaps when a page can be accessed by multiple filter criteria

[Lorem ipsum dolor si]

Hi,

I have designed a real estate site and am having trouble understanding how to do a site map when one page can be accessed by many different urls. The user is presented with a filter form and the url is generated from the filter options chosen, so, as an example, a listing for an apartment could be accessed by

listings/hollywood/listing1234
listings/ca/hollywood/listing1234
listings/apartments/ca/hollywood/listing1234
listings/price[from-100,to-150]/listing1234
/listings/penthouses/listing1234
listings/heated-pools/listing1234

. I could go on.

Is there some recommended way to handle this situation?

[not2easy]

Hello Lorem_ipsum_dolor_si and welcome to WebmasterWorld [webmasterworld.com]

If there is the same content at each of your list of filter results, it is normal to choose which one of those is the primary URL - the one you prefer to have indexed and submit that version in your sitemap. It is preferable to use a canonical metatag to indicate each of the other versions' relationship to that primary URL.

You do not mention what platform your site is based on, but it is not an uncommon situation in a WordPress site.

[Lorem ipsum dolor si]

Hi not2easy,

Many thanks for your response. This is a home-grown website, one I have built from scratch and except for the page with an actual listing, eg listing1234.html, all the search results or filter result urls are virtual except for one which handles responding to each filter criterion - let's call it filter.html which is displayed using an internal rewrite to the url /listings/. So the user then chooses "apartments" which loads filter.html internally rewritten to /listings/apartments/ and which filters out all the other types of real estate. Continuing... user chooses California, which reloads filter.html internally rewritten to /listings/apartments/california/ and the other states are filtered out etc. etc.

If I had to choose an underlying path to a property, I would use the geographical, /listings/apartments/ca/birkley, as opposed to one of the many, many other less direct paths eg. /listings/penthouses/[pets-allowed] or some such indirect path so I would take your advice to mean basing the sitemap on the geographical and, since everything is done at the front end, inserting a canonical url in the inderect paths to point back (assuming the search engine allows dynamically inserting a canonical meta tag).

This also suggests, and I'd appreciate your feedback on this, that let's say should a user find themselves on /listings/penthouses/[pets-allowed]/listing1234, the breadcrumbs would mimic the url, but should the user click on listing1234 without any filter having been applied, that the breadcrumbs would show listings > apartments > ca > birkley.

It's the fact that all the urls are virtual that is giving me a hard time understanding how to suggest to the crawler how the site is to be thought of as being organized.

[not2easy]

Same answer: The actual URL should be the one you submit on your sitemap if the other URLs only exist when generated with the filters. I would add the meta canonical tag to that single page so no matter how the crawler arrived to a virtual URL it would indicate which URL should be indexed.

Google explains why and how: [support.google.com...]

^{PS - I would check your spelling, using maps. ;)}

[edited by: not2easy at 2:09 pm (utc) on Apr 7, 2020]

[Lorem ipsum dolor si]

Many thanks!