| |
|
|
Articles
Dashboards, Analysis Mind Share, and KPIs
They need a dashboard and they need it today. Before looking at any underlying technology or specific solutions, a basic understanding of the types of dashboard and what sorts of things they measure will provide a foundation for any dashboard project.
We see them everywhere; dashboards are another product of the astonishing capabilities of contemporary hardware and software - correct? The dashboard is a powerful metaphor for monitoring business process trends, but like many applications apparently enabled by web technology, it has a long history, almost as long as the automobile dashboard. In France, the tableau de bord is standard on every Citroën and is also the name of a management reporting tool in general use since the mid 1950s.
Tableaux de bord started as manually created graphical (but inert) representations of the state of the enterprise. Web-driven dashboards are dynamic and hyperlinked to sophisticated analytical tools enabling rapid root cause analysis of exceptions.
Dashboards have three well defined categories with different goals and audiences. Before describing these categories, I will introduce the concept of analysis mind share which helps to explain both the appeal of dashboards and puts the various types of dashboards in context. We will also look at guidelines for selecting key performance indicators to populate dashboards.
|
Analysis Mind Share
In a large organization, there is a continuum of responsibility from the
line worker responsible for putting widgets in boxes to the CEO. At both
ends of this organizational hierarchy, the primary functions do not
leave a significant amount of time - mind share - for data analysis
(figure 1). At the bottom of the organization, the focus is on specific
production tasks; at the top, the focus is on strategy. Neither of these
individuals can afford the time to sift through the enormous amounts of
data produced by today's enterprises.
|
Figure 1:
|
 |
Paradoxically, the best interface for these polar opposite groups is
similar - a simple stoplight is typical. If the light is amber or red,
it is time to involve the rest of the continuum - those knowledge
workers in the middle who have the scope, tools, and time to make
detailed analysis not only possible, but frequently a major part of
their job. In the middle levels of the organization, knowledge workers
rely on data analysis to both manage down and report up the
organization.
|
Figure 2:
|
|
In the past the dissemination of results was often performed manually by
those with higher analysis mind share (figure 2). The results were
delivered in written or verbal reports, or for lucky French managers, in
a convenient tableau de bord - dashboard - layout. Business intelligence
tools can automate this process, but this power to easily distill and
distribute summarized information needs to be carefully approached; we
need to understand the different dashboard roles. Dashboards fall into
three well-established categories (figure 3):
|
- strategic dashboards for organizational alignment
- tactical dashboards for measuring progress in projects or
initiatives
- operational dashboards for monitoring specific business
activities
|
Figure 3:
|
|
Strategic Dashboards
Strategic Dashboards measure progress towards strategic objectives; they
help align the organization to strategy in ways that static mission
statements cannot. An executive level dashboard might reflect
enterprise-wide strategic goals and corresponding key performance
indicators or KPIs. A good security model will support enterprise-wide
strategic dashboards that "cascade" down to the department level with
gradually more restrictive views of data, while retaining alignment to
corporate objectives. Working down from global to departmental helps
avoid creating dashboards that pit department against department.
Strategic dashboards are typically highly summarized, highly graphical,
less frequently updated, and include global, external, trend, and growth
measures
Balanced Scorecard Initiative
Strategic dashboards are frequently based on the Balanced Scorecard
methodology of David Norton and Robert Kaplan, a widely adopted method
for determining and achieving organizational goals. Organizations using
this approach are significantly more successful in achieving their
goals.
Kaplan and Norton on their main rationale:
"Financial measures tell the story of past events, an adequate story
for industrial age companies for which investments in long-term
capabilities and customer relationships were not critical for success.
These financial measures are inadequate, however, for guiding and
evaluating the journey that information age companies must make to
create future value through investment in customers, suppliers,
employees, processes, technology, and innovation."
The Balanced Scorecard process gets its name from a balance of financial
and non-financial measures, a balance of short-term and long-term
indicators, and a balance of leading and lagging indicators.
|
A Balanced Scorecard usually contains four perspectives:
|
- capacity for growth
- product development and operations
- customer satisfaction
- financial results
|
Tactical Dashboards
Tactical dashboards measure trends and progress towards strategic
initiatives or special projects, frequently against established goals.
They may involve just one of the four Balanced Scorecard perspectives.
Starting to move away from the stoplight model, tactical dashboards
frequently include summary data as well as visual indicators and make
full use of hyperlinked OLAP tools allowing drill-down and root cause
analysis. The focused nature of tactical dashboards allows more detailed
information to be displayed - the context is clear from the outset.
Incidentally, many dashboards created without a formal approach are
actually tactical dashboards with the aim of maximizing profit or
increasing sales.
Operational Dashboards
Operational dashboards are used to monitor business or manufacturing
processes in near-real-time with the aim of intervening quickly to
resolve issues or take advantage of opportunities. Operational
dashboards are usually departmental in scope and absolute values and
thresholds based on averages and norms are frequently as important as
trends. Like tactical dashboards, the focused nature of operational
dashboards allows more detailed information to be displayed. It would be
unusual for a top level manager to use an operational dashboard; a
traffic light summarizing operational capacity trends would be more
appropriate.
Key Performance Indicators (KPIs)
Effective dashboards are composed of key performance indicators. An
organization may have many viable trend or performance indicators; with
limited dashboard real estate, we must identify the key performance
indicators (KPIs). Starting with organizational goals, we evaluate the
measures of behavior and events available to ensure they are both key
and performance-related.
KPI candidates must be key measures, crucial to business strategy and
they must link to performance. There should be a cause-and-effect
relationship between actions and indicator. They should clearly
distinguish between effective and ineffective performance. Employees
must clearly understand and have control over achievement of the KPI.
Order fill rate and cycle time are typical KPIs - they neatly summarize
a number of measurements, are easy to understand and target, and have a
positive impact on customer satisfaction.
A Dashboard Solution
There are a number of dashboard solutions on the market; drawing up
requirements based on analysis mind share, dashboard type, and KPIs will
help narrow the list. The tolerance for analysis will drive the
graphical and numeric content of a dashboard. A clear picture of the
types of dashboard your organization requires will help define where
your solution lies between displaying cached results for a strategic
dashboard and interactive updates for some operational dashboards. The
nature of your key performance indicators will help you understand how
much pre-aggregation is required before summary data can be displayed.
This should launch you on the way with what, by definition, is a
high-visibility project. |
|
|
|
|
|
About the Author:
 Ross
Morrissey is a Product Specialist at Management Information Tools,
Inc. developers of MITS Advanced Reporting and Business Intelligence
products. Ross is the designer of the MITS Dashboard Training Curriculum
and has discussed Operational Dashboards in educational sessions at
International Spectrum and MultiValue User Groups over the past 5 years.
You can contact Ross at (888) 700-MITS or
rossm@mits.com.
This Dashboards, Analysis Mind Share, and KPIs white paper was
Spectrum Magazine's cover article for the September / October 2005
edition. |
|
|
|
