Using the example of the innovation of a customised home page from the early days of Flickr, this article helps break down how to delight users:

Years ago, we came across the work of Noriaka Kano, a Japanese expert in customer satisfaction and quality management. In studying his writing, we learned about a model he created in the 1980s, known as the Kano Model.
The article does a great job of explaining how you can implement great features but they don't particularly get users excited:
Capabilities that users expect will frustrate those users when they don’t work. However, when they work well, they don’t delight those users. A basic expectation, at best, can reach a neutral satisfaction a point where it, in essence, becomes invisible to the user.

Try as it might, Google’s development team can only reduce the file-save problems to the point of it working 100% of the time. However, users will never say, “Google Docs is an awesome product because it saves my documents so well.” They just expect files to always be saved correctly.

So it’s a process of continual improvement, and marginal gains in some areas:

One of the predictions that the Kano Model makes is that once customers become accustomed to excitement generator features, those features are not as delightful. The features initially become part of the performance payoff and then eventually migrate to basic expectations.
Lots to think about here, particularly with Project MoodleNet.

Source: UIE