How do the Quoras of this world index their content?

[freek27]

I am helping a client index lots and lots of pages, more than one million pages. They can be seen as questions on Quora. In the Quora case, users are often looking for the answer on a specific question, nothing else.

On Quora there is a structure setup on the homepage to let the spiders in. But I think mostly it is done with a lot of sitemaps and internal linking in relevancy terms and nothing else... Correct? Or am I missing something?

I am going to index about a million question and answers, just like Quora. Now I have a hard time dealing with structuring these questions without just doing it for the search engines. Because nobody cares about structuring these questions. The user is interested in related questions and/or popular questions, so I want to structure them in that way too.

This way every question page will be in the sitemap, but not all questions will have links from other question pages linking to them. These questions are super longtail and the idea is that when somebody searches this exact question we can supply them with the answer (onpage will be perfectly optimised for people searching this question). Competition is super low because it is all unique user generated content.

I think best is just to put them in sitemaps and use an internal linking algorithm to make the popular and related questions rank better. I could even make sure every question has at least one other page linking to it, thoughts?

Moz, do you think when publishing one million pages with quality Q/A pages, this strategy is enough to index them and to rank for the question searches? Or do I need to design a structure around it so it will all be crawled and each question will also receive at least one link from a "category" page.

[freek27]

Wow, that is insane right?

https://www.quora.com/sitemap/questions?page_id=50

I wonder how long this carries on.

[Bigb06]

Quora don't seem to have a XML sitemap but a HTML one :

[https://www.quora.com/robots.txt](https://www.quora.com/robots.txt) refers to [https://www.quora.com/sitemap](https://www.quora.com/sitemap)

[freek27]

Yes there are many challenges and external linking is definitely one of them.

What do you think about sitemaps to get this longtail indexed? I think that a lot can be indexed by submitting the sitemaps.

[EGOL]

There are many challenges to building a really large site.  Most of them are related to building the site, but one that often kills the success of the site is the ability to get the pages into the index and keep them there.  This requires a steady flow of spiders into the deepest pages of the site.  If you don't have continuous and repetitive spider flow the pages will be indexed, but then forgotten, before the spiders return.

An effective way to get deep spidering is have powerful links permanently connected to many deep hub pages throughout the site.  This produces a flow of spiders into the site and forces them to chew their way out, indexing pages as they go.   These links must be powerful or the spiders will index a couple of pages and die.  These links must be permanent because if they are removed the flow of spiders will stop and pages in the index will be forgotten.

The goal of the hub pages is to create spider webs through the site that allow spiders to index all of the pages on short link paths, rather than requiring the spiders to crawl through long tunnels of many consecutive links to get everything indexed.

Lots of people can build a big site, but only some of those people have the resources to get the powerful, permanent links that are required to get the pages indexed and keep them in the index.   You can't rely on internal links alone for the powerful, permanent links because most spiders that enter any site come from external sources rather than spontaneously springing up deep in the bowels of your website.