You know when something gets flung about so often that it loses all meaning? Or maybe, you’ve just heard the phrase so many times, you’ve simply assumed its meaning without ever having actually stopped to consider it. Business Intelligence
are one of those ‘things’.
Orwell on BI dashboards
Famous linguist, George Orwell
, noted and lamented this emerging pattern within the modern English language – the tendency to use metaphors and well-worn phrases to express oneself at the expense of conveying meaning via carefully considered, original, context-appropriate and structured communication.
Orwell contended that “the sole aim of a metaphor is to call up a visual image”, and that we, as humans capable of communicating to one another, should “never use a metaphor
, simile, or other figure of speech which you are used to seeing in print.”
Whilst this lofty standard might be a little hard to wholly adhere to, it’s worth considering Orwell’s position that, contrary to its intended purpose, “stale metaphors hamper our means of expressing ourselves”.
Orwell was suggesting that the drabness of the overused metaphor actually removes meaning – it makes our prose less precise. In other words (Orwell’s words in fact), metaphor-riddled prose “consists less and less of words
chosen for the sake of their meaning, and more and more of phrases
tacked together like the sections of a prefabricated hen-house.” In short, we don’t say what we mean. We let easily conjured words and speech patterns construct a sentence for us. The result is language which is neither specific to the particular situation under discussion, or transfers connotation, value or understanding clearly. As Orwell said, “… the worst thing one can do with words is surrender to them.”
It seems very plausible that BI dashboards
are suffering from a comparable state of lethargy. A persistent climate where the purpose of a BI dashboard, as a device used for communicating organizational data from machine to man (and hopefully helping to turn that information into knowledge and then action), has been forgotten.
Analytical design and data visualization
proponents, Edward Tufte
(described as “The Leonardo da Vinci of data” by the New York Times) and Stephen Few
, have both delivered dismal assessments of the modern BI dashboard, arguing that most exponents, products and developers fail to build, communicate and deliver successful dashboards because they don’t:
- Think about how the individual reports within a dashboard interrelate;
- Consider the how and the who of information consumption, and most importantly;
- Acknowledge its intended purpose – to convey information quickly and unmistakably.
As Orwell submitted, if one person writes an essay using clichéd thoughtless phrases, and another writes one with prudently selected vocabulary and expression, both have produced an essay. However only one person has produced something worth reading. Likewise, a BI dashboard constructed in ignorance or abstinence from communication best practice cannot deliver insight and expose the true value of an organizations’ data.
Ironically, dashboards themselves – by name and by nature – suffer from being a metaphor. Maybe that too is part of the reason why people become distracted with other superfluous aspects of ill-advised dashboard design? Perhaps because people are constructing a piece of communication within a metaphor, they forget its intended purpose?
For more on the ailments of the modern dashboard, visit our provocative blog post, Are Business Intelligence dashboards on the brink of extinction?
However, despite the convictions and apparent pessimism of both Tufte and Few, it seems that BI specialists, at least many of them anyway, understand the fundamental purpose of the BI dashboard. In a shrewd discussion thread in LinkedIn Group Business Intelligence Professionals
, various experts and authorities on BI discussed the “Difference Between Balanced Score Card (BSC) and Dashboards
Gary Smith, a self-described expert in Project Management and Process Excellence, defined a dashboard as “the graphical display of the KPIs [Key Performance Indicators] and other metrics for management. Data must be collected and aggregated for programmers to develop the dashboard.”
Douglas J. Roach waded into the discussion with this contribution: “A dashboard is a management and decision tool that is based on data that is as real-time as possible. It is meant, like your automobile dashboard, to give you a clear indication of where you are headed, how fast, and the health of your business engine. It may be focused on one aspect, such as security, or on project health and progress; but it should provide indications when something seems wrong (based on clearly defined and documented metrics), and the best ones allow a drill down to the level where the trouble(s) originate.”
Meanwhile, BI consultant Chris Gerrard despaired, stoking the Few-Tufte fire (sorry George):
“So many people with so little knowledge… There are opinions and facts. It’s good to know the difference.”
We couldn’t agree more Chris. So, to get all the facts about BI dashboard best practices, join us for our upcoming Orwellian Webinar Event >
Perhaps the proliferating problem of dashboard confusion and confabulation lies in the consumerization of BI – where many more people, from wide ranging and quite often non-analytical backgrounds, are now utilizing and being exposed to the concepts of dashboards and data visualization via reporting and analytics
This potential explanation may simultaneously elucidate the growing misuse or misconstruction of dashboards (either by a less informed audience or to appeal to the more visually gratuitous demands of that audience), while also alluding to the growing importance of adhering to dashboard design, delivery and communication best practices in order to convey the clear meaning of data to an unfamiliar audience.
What do you think drives people towards feebly conceived dashboards? Poor planning? No clearly defined reporting objectives? Misunderstanding of dashboard design and delivery purposes or capabilities?
Let us know at email@example.com