There is a very important new concept in SharePoint 2013 publishing: search driven publishing. This is a concept that breaks down site collection boundaries (get search index content from multiple site collections), allows for flexible and dynamic publishing, separates presentation from storage and eliminates large list thresholds. The existing publishing model in SP2010 still works, but the new search driven capabilities gives you great new possibilities.
Using the Content Search Web Part, Catalogs, Managed Navigation, Refiners, Usage Analytics stored in the search index, etc., gives you a truly search driven publishing web.
Content Search Web Part
The main idea behind search driven publishing is to use the content of the search index (the crawled content) to publish, republish and target information to the end user. The single most important part of this functionality is the new Content Search Web Part. This is a web part that enables us to build queries in the UI (the Query Builder) to show content, use Query Rules to further manipulate the search result (for example targeting, related content, etc.). The web part can be added to any page and can be found in the “Content Rollup” category, along with the good old Content Query web part:
One “problem” when using this web part is that it is based on the index, and there is an index lag – but then again, the new enhanced search capabilities of SharePoint 2013 (for example continuous search indexing) should minimize this problem.
Another important note to make about the Content Search Web Part is that it probably not will be available in Office 365 (but this has not been definitely stated at the SharePoint conference as far as I know).
Content Catalogs, Managed Navigation and Refiners
The concept of search driven publishing also includes the new Catalog concept of SharePoint 2013. Any list or library can be registered as a catalog, and page templates can be built for items in a catalog. A typical usage is a product catalog where there is a product page template, but it can also be used for any other type of content (for example news, articles, list of helpdesk tasks, etc., etc.). Using managed metadata from the taxonomy term stores enables you to build a navigation based on terms. So for instance, if all products in your product catalog is tagged with a product category, this can be used for navigation.
Another extremely useful implementation of using managed metadata is the possibility of showing Refiners for a list of items in a catalog. Refiners are used to filter down the items shown on a category page. An example could be that you refine a product catalog page to show only products that have a certain color, is in a certain price range, etc.
Another very important part of search driven publishing is the new capabilities of usage analytics in combination with search. The new Analytics processing engine, in addition to be used for analytics reporting, now actually stores the data in the search index! This is very good news: you can now easily use the Content Search Web Part to show recommendations (“users who bought this also bought…”), target content, etc. This functionality uses so called Event types stored in the Event Store, and there are many predefined event types (for example “ViewsRecent” for recent viewed pages). It is also possible to define custom event types that will be stored in the Event Store and can be used along with the built-in types.