Magnolia is often seen as a simple to use system for maintaining web site content. It is very easy to generate html (the most obvious output) and to edit content within the site structure. But that is not all of it.
There are often overlooked aspects to Magnolia's publishing abilities. I always like to point out that Magnolia's templating architecture is output-agnostic. That means that we can publish anything, not just HTML. This is supported through various means, one being our ability to map url-postfixes to sub-templates. For example, try .rss, .css or .print instead of .html for the magnolia.info home page to see othr publishing formats.
And while it is true that our standard interface is very easy to use because it is based on the structure of the final web site, we also have abilities for content reuse, in fact decoupling the site structure from content. New in Magnolia 3.1 we will finally add the data module, which takes this to a new level, allowing you to manage structured data in Magnolia. We can consume (and produce) RSS. This allows you to maintain your content outside of a site structure and populate the page content with that "external" (to the page) content.
Finally, the support of selectors allows you to access parts of the rendered page if your templates support it. Example: you can just get the header, footer, body or left column of a page if you want.
Anyways, there is much more to Magnolia than meets the eye of a casual browser, which is why we always recommend our Subject Matter Experts to be part of large scale projects.