We use cookies and other similar technologies (Cookies) to enhance your experience and to provide you with relevant content and ads. By using our website, you are agreeing to the use of Cookies. You can change your settings at any time. Cookie Policy.
Start a Discussion
BI
user-pic

Are Dashboards the Solution to 'Too Much Information'?

Vote 0 Votes
There has been plenty of discussion lately on the strengths and weaknesses of dashboards (see Ted Cuzzillo's blog). So are dashboards the answer to 'too much information,' or are they too much of a simplification?

11 Replies

| Add a Reply
  • A well designed set of dashboards will have visualization, text and numeric data that will bring focus to key performance areas that need to be managed.

    Dashboards should have graphics that convey information and do not clutter or confuse the consumer with overly ornate or meaningless visual filigree. Effective dashboards have context and provide facts, trends and comments that simplifies the consumption of information, hopefully “reducing information overload.�

    Chase in point, Napoleon. What if Mr. Bonaparte just picked up his new iPad in 1812 with his new world domination informational dashboard? Napoleon configured his supply chain and supply line data with spark charts designed by Stephen Fey. A “mash-up� map overlay from the Weather Channel provided forecast temperatures for the next 60 days. Dominating the Emperor’s tab front and center was a daily update to the famous “flow� chart by Charles Minard depicting the disastrous attrition of his army in the 1812 campaign into Russia. Who knows, it might have been information overload, not just arrogance that spelled his defeat but alas, the iPad still doesn’t support CDMA.


  • I am surprised at the claim that Dashboards contribute to too much information. If anything, they declutter information and provide a high level view where the numbers start to make sense.

    Companies have been having at their disposal numbers - lots of numbers. But what they lack is the higher level view of what it means. Increasing? Decreasing? What is?
    Slowly? Fast? How do the numbers compare with goals? How are the averages made up - which ones contribute to the increase vs which ones the decrease? Bad performers cloud good performers in averages.

    Well-designed dashboards are essential in providing a summary view of what the numbers mean.

    Of course, as with anything, abuse of dashboards should not cloud their intended purpose. If well done, dashboards could become invaluable in teasing out and highlighting essential information!

    My two cents!

  • Dashboards are a great way to present information as long as:
    1) Its User specific (Business relevant)
    2) Information presented are Key indicators
    3) They can be configured
    4) Properly and timely channeled – Critical (Hand held – targeted dashboard) or Alerts(emailed/sms – non-dashboard) – its less likely that users will login to a web based application and keep looking at the screen
    5) Its contextually relevant - A user may need different information at different times – daily, weekly, monthly
    6) The data source is correct
    Rest all is ornamentation!

  • I like the above replies but would like to take it further. The question is to what purpose do you use a dashboard? It is not just to provide relevant information out of too much data - it is to enable people to make better decsions more quickly by using actionable information and analytic capabilities. This means the dashboard is simply the front-end interface that supports a continuous improvement or decision cycle.

    To be effective dashboards must present timely information in context and against targets, but they must also:
    - provide alerts highlighting issues (these should also be emailed to designated people);
    - provide guidance in how to interpret the information on the dashboard;
    - allow people to share information and make annotations on specific data points that are anomalies or of interest;
    - allow people to drill into the drivers behind the performance indicators;
    - help people characterize issues with easy to use statistical algorithms such as trend lines, forecast projections, control charts, variance analysis, histograms, Pareto charts etc;
    - enable people to perform "what if" scenario analysis to determine the most appropriate action to take to resolve an issue or improve performance.

    In this way the dashboard becomes an interactive tool to enable people to make better decisions and not just a convenient way to visually show information.

  • Very good replies so far. I'll just add that dashboards can surely help cut through the data clutter and summarize the most meaningful information and indicators through an appealing interface, pre-defined drill paths (best practice-based) and drill-through to the right source data elements where possible and necessary. I can tell you first hand that, done properly, a great set of dashboards become the operational lifeblood of an executive's role. I'll add that creating a great dashboard is a mixture of art and science. The science is described in several of the past responses and the art is mentioned in the first (Dyke Hensen's reply). The notion of adding contextually-relevant information from external sources (usually web-based), alongside data-driven charts, gauges and indicators is a valuable way to make a dashboard more decision-focused and action-oriented. I see customer projects emphasizing this element more and more all the time.

    Brian Gentile
    Chief Executive Officer
    Jaspersoft

  • Dashboards, like all tools and technologies, are useful when they are used appropriately. Dashboards are ideal for presenting dynamically changing or cumulative information based on numeric source data. Some examples of useful information include understanding the utilisation of IT systems and tools, throughput from automation or manually entered data. The information provided is a key input into root cause analysis and is invaluable in providing quick 'health check' views of the systems which comprise an organisation.

    Dashboards are dependent on the ability to measure the value or values of some function. For dysfunctional organisations where systems, groups and functions aren't well integrated then dashboards are less useful.

    A well designed dashboard can provide an insight into the emergent properties that arise from connecting systems. This is, in my experience, the true value of dashboards. Realising value and continuous improvement are the high level benefits of dashboards.

  • Great answers! Especially the one by Dyke Hensen on Napoleon's iPad made great reading.

    The question reminds me of the blog post I had written sometime back, on the subject of data visualization, titled "Michelangelo and Da Vinci in your BI Team". (http://www.beyeblogs.com/karthikonbi/archive/2010/01/).

    Dashboards,if created properly, are a great way of rendering actionable information to business users. Having said that, here are some points to consider:

    1) Dashboards should be 'cascadable'. For this, they should be created based on an organization's strategy map (or some such technique) that shows the inter-relationships between the metrics.

    2) An exercise in profiling the metrics (similar to data profiling) before dashboard creation is immensely useful.

    3) It is probably worthwhile to remember that dashboards though useful cannot communicate the non-linearity associated with business processes. In my humble opinion, a combination of dashboards with business process simulation models would make for a killer app.

  • Good thread. The question lead me to think what is the definition or nature of a dashboard is (as a delivery mechanism)

    "A dashboard is a visual presentation style on a collection of vital information required to achieve one or more objectives/goals."

    This presentation style should highlight the what or pattern in the information rather then revealing the "why" its happening.

    Further analysis does require the detail; the dashboard can be customized/configured as this starting point to drill down into the data; it can also serve as just a high-level summary.

    The essentials really being concise, clear, at-a-glance summary, with some intuitive flow to the information (as describe by Dyke Hensen) that can be used to effectively state the required message.

  • Done well, dashboards are clearly the best answer to ‘too much information’. Dashboards are sign posts to areas that need deeper investigation. In this sense, simplification is a very good thing. Given the information overhead knowledge workers face, there has to be some sort of early warning system to help point focus to the areas most in need of attention. Based on the way most of us process information, especially younger generations, visual cues are the most effective means to accomplish this. Of course, that being said, great care must be taken in setting up the dashboards to insure the right KPIs for each user type are tracked so important measures don’t slip through the cracks!

  • Hmm

    I think Dashboards can be part of the solution but I am not sure they are the best way to cope with the flood of information deluging companies.

    I think the reality of many companies is that all too often the person who needs to exploit the information (of which there is too much) is not analytically sophisticated or even that there is no-one watching, as in an automated system. In those cases a dashboard needs to be reporting on what was done (and how well it worked) while the information flood is turned directly into executable analytics so systems can act or advise directly. In this case the dashboard is helping us understand how our systems handled the information not helping us understand the information directly.
    JT

Add a Reply

Recently Commented On

Monthly Archives

Blogs

ADVERTISEMENT