Monday, June 3, 2013

Reports are Dumb

Really, reports are dumb. This is not to say that people that use them are dumb, not at all. This is really about the fact that reports, as we have come to know them, don't do much for us or our business partners.We can make some seriously pretty data output with Crystal Reports, Business Objects Webi, or SQL Server Report Builder (well, maybe not that last one) that answer key business questions. That sounds important, and it is, but the problem is that the usefulness of the report ends with the answer.

The Hitchhiker's Guide to the Galaxy (book/film) imparted a vital piece of wisdom. It isn't enough to know the answer to a question or problem. We need to understand the question as well. It might even be better to know the right questions to ask.

I would like to see IT help their business partners take this concept a step further.Why not give our users something that not only answers their question, as typical reports do, but leads them to the next question or problem.Why not give them the tools and information that lets them look at the data and ask "Why?" Why do they get the answer they see on the screen. Why do the get an answer they don't expect? Give them the power to analyze their data in depth and from different angles.

In case you haven't figured it out already, "self-service BI" is the direction I am headed. This can be a scary space to enter into and it is not without its pitfalls. A lot of thought and work goes into building a self-service BI infrastructure but, in the end, your company (or customer) will be a lot better off for it.

We sometimes think of our business partners as children. We give them toys and they find a way to abuse them and sometimes hurt each other. However, if we give them a little guidance and establish some boundaries it's amazing what games and other ideas they will come up with.

This post was inspired by the keynote and the Big Data, Small Data, All Data session presented by Shawn Bice, et. al. at TechEd North America 2013. If you have something to share about this article or something from TechEd please post a comment or link to your blog post.


  1. I disagree with "the usefulness of the report ends w/ the answer." The report drives action, and helps validate if the action taken worked. If not, try another type of action and check the report again. IE: adoption of a new feature or product.

  2. Thanks for the comment Anon, you do raise a valid point. I would argue that the report doesn't drive the action though, the data does.

    If the marketing folks and/or project managers had a report that tells them how well a new feature is being adopted would that be good enough? Wouldn't it better if they had access to the feature usage data and could drill in and find out why the new features were or were not being used?

    The report limits them and keeps them from possibly using their imaginations to find out more about what is going on. That is the strength of self-service.