Database Reference
In-Depth Information
A quICk unDeRSTAnDIng:
glAnCe AnD go
The main reason for implementing a scorecard system is to enable quick
analysis of key metrics—the ability to view a metric (be it a KPI, an operational
metric, or indeed any metric in your business) and either move on when values
are within predefined bands, or start an investigation when the values are
out of band. This definition often sits outside the traditional scorecard and
dashboards arena, but is a very important and growing area for these tools.
Some key design features to enable this are vital—the details vary from tool
to tool, but the basics are the same:
uu
A filter area to change the data displayed
uu
A scorecard area, with drill-down capabilities and indicators, to highlight
where investigation is needed
uu
A further details area—either drill across or drill through
KPIs
KPIs are the heart and soul of scorecards. In its simplest form, a KPI is an actual
value, measured against a target, and assigned to a band by the distance from
the target. The reality of choosing KPIs for an organization's strategy is that
the process is long, complicated and fraught with difficulties—for instance,
measuring customer service representatives by how long they spend on the
phone with customers (what appears to be an easy way to drive down costs)
can have grave effects.
In this topic, it's assumed that you've gotten all of that perfect, and you need
to master using the technology.
The other piece of this puzzle is that technologies allow you to do much
more than simply assign an actual, a target, and bands to a KPI. For instance,
in PerformancePoint you can add both dimension and time filters, and have
multiple actuals and targets—for example, actuals of Today, Month-to-Date,
Year-to-Date, with targets of This-Day-Last-Year, Last-Month-to-Date, and
Last-Year-to-Date. The bands are the range of values that are considered
good, acceptable and poor. For instance, 120 percent and above against the
previous year would be good, anything above 100 percent of the previous
