Case study: User driven data resource creation

A large organization found that the information needs of their various knowledge workers in sales, marketing an operations are varied and quickly evolving. They recognized that full blown IT projects are too heavy and slow to satisfy the constantly changing needs for particular, easy to use data.

For example, a senior member in their sales team finds a new metric by which to judge sales success probabilities. However, to retrieve a list of sales prospects - sorted by this new metric - either requires time consuming manual work, spread-sheet magic on painfully extracted customer tables or a formal request for IT to implement a new data service. The proper data service of course is preferable: One place to repeatedly go to in order to get an automatically updated list. But IT is busy, the new request is one of many and delivery of the new service may be weeks or months away. In the meantime, sales opportunities are lost.

The organization wants to become more agile: Knowledge workers and users need to be able to create and share data on their own. What if the senior sales consultant would be able to take the knowledge about how to sort a list of prospects by that new metric and create a new data service themselves, without having to involve IT? Traditional Wikis have failed them: Users can create new data, but it's static, consumable only by people. What's needed here is the creation of a URI, something other users in the enterprise can just click on or import into their spreadsheets, but which is always automatically updated.

RESTx was the solution: It allowed the IT department to encapsulate their key data assets in a few components: Some of them off-the-shelf (MySQL database, Salesforce), some of the custom code (a proprietary CRM system, legacy databases). After that was done, users and knowledge workers were able to quickly and easily create new information and data resources based on those components, by providing custom configurations (such as queries or parameters) and making these available as new resources. The data that is provided by those resources is always fresh and is suitable not only for humans to look at but also fit for import into spreadsheets, mashups and other client applications.

The data represents a different and easy to use view of the existing data sources in the enterprise, allowing read and write access straight back into the original data sources where deemed necessary. Best of all: IT does not need to get involved in the creation of any of these custom resources. The users themselves can quickly and easily create new ones.

The senior sales consultant provided a new configuration for this 'SalesProspect' component merely by filling out a form in the web browser, providing particular query parameters to sort the sales prospects. Since this takes only seconds, the sales consultant was able to experiment with a few variations on those parameters and even found an improvement on the metric. Once satisfied, the URI of the new resources was shared amongst the other members of the sales team. For them, this URI is short and opaque: Just a simple handle to retrieve a customer list, sorted by a particularly interesting metric.

RESTx automatically provides an easily browsable registry of all resources, along with documentation on how to use each resource, it's possible parameters, and so on. Therefore, there is no extra task of maintaining such a registry. All information about RESTx resources and components can be reached by following links from the server's root directory. Since the organization utilized an enterprise search engine, it was quickly possible for everyone to conveniently find appropriate resources for a given task merely be looking for a few keywords in through their enterprise search.