Untitled Document
Key Performance Indicators provide enterprises with the much needed insight
to understand the health of their organization and enable them take necessary
steps to overcome the bottlenecks. Traditionally, the source of data for these
KPIs has been either from a data warehouse or a data mart reflecting the concept
of traditional BI as discussed in my previous article, Operational
vs. Traditional BI: Context Matters.
From a strategic point of view it may have sufficed where the involvement mainly
has been by the top executives and thus the KPIs have been more aligned to reflect
high level metrics required from a strategy perspective.
With Operational BI gaining acceptance by enterprises as they are beginning
to understand how the organization as a whole and individuals at various levels
can help the organization to be more successful, the KPIs need to also reflect
the same thought process. Say a planner (manufacturing) would have a KPI such
as "requested date of arrival of goods" based on which they will provide
the production schedule to the shop floor.
A line of business representative, who is directly dealing with the customer,
would have a KPI such as "goods delivered on time (by shipment #),"
"delivery delayed." On the other hand, the finance would probably
have a KPI something like "shipment delivery note received" and "payment
received (by shipment/order)" -- this will help them to raise the invoice
and track the payment. From a top management perspective they could then use
all these KPIs to create their own set of KPIs to understand the current business
situation and act upon it.
Here, KPIs can be rolled up just like services from an SOA perspective thus
highlighting the fact that KPIs can have granularity (business KPI) and will
be more effective. This would reflect a realistic hierarchical KPI tree representing
the kind of information an organizational hierarchy is interested in.
Also, the data in question would be more up to date as it would reflect transactional
data from the data source applications themselves, though it would be important
to understand how the data is acquired from these disparate applications. Here,
I strongly feel SOA will play a key role in fetching the data in a standardized
format and possibly use service definitions to standardize the KPI format and
define them.
1
Print this article
Email this article
Talk Back!
Write to Editor
