Rethinking Reporting
I have been doing a lot of briefings for the "Magic Quadrant for E-Recruitment Software" update. One of the things that has stood out to me is the lack of innovation (for the most part) around reporting and analytics. Most vendors have three components to their reporting/analytics offering:
- Standard Reports -- These include both regulatory-related reports as well as common reports that most recruiters, and to a lesser extent, hiring managers want to see.
- Ad-Hoc Reports -- All of the vendors I have talked to thus far have an ad-hoc report writer that allows the vendor or the customer to create their own reports beyond the standard ones. The standard reports are usually created with the ad-hoc reporting tool.
- Dashboards -- Many of the vendors offer some sort of role-based dashboard that provides what they vendor believes are key metrics for the role. The metrics are configurable so if a customer wants different ones on the dashboard for a particular role (or in some cases the user can change it themselves), they can make the change.
The challenge with the offerings is that it puts the onus on the user to find the information that they truly need. A user has to go into the list of standard reports and find which one will answer the question. If that does not work, then they need to create a new one in the ad-hoc reporting tool. Similarly, if a user is lucky, a dashboard will provide the information that they need to answer a question (or the ability drill-down into metrics will provide the information).
Given what we have learned from the world of search, why does it still work this way? Why can't a user just enter a question and have the system suggest which reports, graphics, or dashboards would best answer the question? I am not suggesting Natural Language Processing (NLP). I am thinking about an approach akin to a Google search. Google does not provide "the answer". It provides links to what it thinks best matches what the user asked for. The user can navigate the suggestions and choose which one will work best. It can work the same way for reporting. Even if the "best fit" report suggested does not quite provide the answer, at least it provides the starting point for using the ad-hoc reporting tool to answer the question.
An HCM application also provides context for the search. So, if I am a hiring manager and I am creating a requisition, then there may be a set of common questions that the system could suggest (e.g., what are the characteristics of high performers in this position? what is the market rate of pay for this position relative to our budget?). If you ask a vendor during a demonstration, "How long does it typically take to fill a requisition for this position?". They can easily navigate to the "Time to Fill" report and then drill down into the specific position data. They should be able to do it -- they are experts on how to use the reporting tools. You can get training so you can do it too, but why should you? Since the vendor knows the best way to answer the question, why not build that intelligence into the system (and have the system learn as users choose which reports best answer specific questions)? If I ask the same question multiple times, then maybe the system can save the question and the report I typically use to answer the question as a "favorite".
In this age of Web 2.0, there is a lot of focus on improving the user experience. In reporting, it has focused on providing more graphical views of information and drill-down. What do you think? Are vendors doing enough to provide users the information they need to make decisions and take appropriate actions?
