« Swat versus Clobber: getting and using the Upper Hand | Main | Making Advantage Sustainable »

December 3, 2005

Performance Analysis and the Six Types of Performance Management Information

Performance is a measure of the degree to which observed effects are meeting the plan.

In management, planning represents the "agreement" that a manager has proposed between the possible and the probable.

I.

The plan is developed from a theory or model of how progress towards value results from the organization of the operations.

But a more general function of the model is to define all conditions that must be present for value generation, and describe why their combination is coherent for the purpose at hand. That is, the model describes conditions that drive progress, while explaining why those conditions must be supported.

Meanwhile, monitors will determine that a condition is in this state or that state, while the plan describes the target state desired. That is, the measurements supported by monitoring presume that conditions and states are two different things. Observation confirms conditions, while measurement confirms states. (For example, cooking requires heat -- representing a necesary condition; the heat may be at a certain temperature, representing a state.)

This means that managers must not only work on distinguishing conditions that matter from those that don't (as per models), but also the states (of conditions) that matter versus states that don't (as per plans). Only within a certain range of states is a condition significant, and only at the proper state are certain conditions significant.

II.

Management attends to the actual versus the planned organizational status of operations.

By "organizational", we mean the distribution of resources in a structure designed to support fulfillment of demand. This structure is an implementation of the model, in the same way that an actual building is an implementation of the design in a blueprint.

In management, we assume that the necessary conditions are in the required states unless proved otherwise. If things are proved to be otherwise, we revisit the necessity of those conditions, the states of the conditions, or both. Faced with gaps between the actual and the planned, we make repairs or we make modifications -- to restore coherence to the conditions, restore compliance to the target states, or both. But to do that appropriately and tolerably, we have to account for our decisions with information that
- (first) describes and confirms the gap, and
- (second) interprets the gap for its probable meaning versus our purpose.

This all amounts to six uses of information that contribute to a full cycle of analysis about performance. Information is used to:
1 - develop the model
2 - develop the plan
3 - initialize and prioritize action
4 - monitor status
5 - confirm underlying and correlated events
6 - reconcile differences.

With those six uses of information, managers can continually influence the state of conditions that are likely prerequisite to the capture and delivery of value to the organization's targeted stakeholders.

III.

The purpose of organization is to support desired behavior.

In managing performance, the desired behavior has significance because of the effects it produces relevant to the model.

The first priority of behavior is to create the states that conditions need to be valid per the model. Variance from targeted behavior is not just an enforcement issue, but since success sometimes accompanies variance it is also a way of understanding that different but similar plans could work for the organization immediately.

Many different behaviors may achieve the same effect, so if there is any reason to prefer one behavior over another, analysis of the behavioral facts is key to finding the best behavior. Because the main point is just to distinguish the behaviors from each other, these "facts" may be more descriptions than measurements.

In producing these descriptions, most managers are actually focused on "business intelligence" (BI) instead of on performance analysis. This is because the Plan is the central and most public manifestation of management. Business intelligence has the capability to expose facts such as patterns of relationships, that can then be judiciously embraced or avoided for the future. By suggesting the needs or options to make decisions that affect compliance with the plan, BI broadcasts visibility of the accountability of managers.

Meanwhile, as an aftereffect of increasing exposure, BI can have the "feedback" effect of informing and suggesting evolution of the organization's theory of progress, so modifications to performance plans might also follow.

But performance analysis directly concerns itself with the degree of tolerance the organization has for complying with its own model. In that way, performance analysis will explore how flexibly the plan can be executed before the plan is considered "violated" or "broken"... and thus unable to support the ultimate goal for which it was designed to serve.

IV.

Performance managers must recognize that performance analysis must be done with or without BI. In order to understand how that can make sense, managers need to understand that while observation and measurement both generate descriptions, they are not the same thing.

The essential purpose of BI is to gather enough evidence of activity patterns, in operations and transactions, for identifying the most probable patterns associated with "valuable" outcomes. However, BI starts without an idea of why things work, and builds a picture of what is working or not, regardless of the plan.

In comparison, the main purpose of performance analysis is to discover and prioritize the factors having the most significant current impact on the "effects" of the plan's execution. Performance analysis information starts with the existing idea of why things should work, and acquires more information to test how to make it work in light of reality.

Business intelligence clearly has a role in performance analysis, because it can help tremendously with the discovery tasks in performance analysis. However, BI's capability to help is not automatically used that way. As is pointed out by many experts: "it's all good for measuring and reporting information from events, but if you don't have a hypothesis to test, you can be incredibly efficient in measuring events and still learn nothing about how things really work." (Christopher Kenton, columnist for Business Week Online)

Furthermore, representing the hypothesis and how things really work is a task for "knowledge" as opposed to "information". The role of an information system is to feed knowledge, but then management needs a knowledge system to process the information.

In the knowledge system, here are six fundamentals for the performance manager:
1 - Goals
2 - Assumptions
3 - Progress
4 - Priority
5 - Variance
6 - Status

Again, the objective of performance management is to identify and control occasions calling for repair or modification of the coherence of, or the compliance to, conditions prerequisite for appropriately meeting demand. Performance analysis features a method for interpreting information pertinent to that objective.

V.

No less impressive, but different, is BI. In this definition from an e-commerce glossary of the Pennsylvania Department of Environmental Protection, we see the difference and the overlap:

Business Intelligence (BI): An interactive process of analyzing and exploring structured, domain-specific information (often stored in a data warehouse) to discern trends or patterns, thereby deriving insights and drawing conclusions. The BI process includes communicating findings and effecting change. BI domains include customers, products, services or competitors.

That notion of BI "process" is a little hyperbolic; it really is describing the uses of BI that give it importance, rather than the process of BI er se. However, it is definitely appropriate to put BI into the larger perspective of business management. Along that line, we consider that BI can affect the perception and acceptance of the business model:

Business model: a design of the operations of a business which focuses on how revenue will be generated. -- Webster's New Millennium™ Dictionary of English

BI's emphasis on domains is vital to validating the integrity of "business models". It tests assumptions about processes and relationships that underlie expectations of operational output.

But performance analysis' emphasis on behaviors is vital to validating models of business value and operational outcomes. This takes the management consideration to a higher level, for understanding the organization's capability, relevance and fit to its environment instead of just its activity and discipline.

Business value: In management, business value is an informal term that includes all forms of value that determine the health and well-being of the firm in the long-run. Business value expands the concept of value of the firm beyond economic value (also known as economic profit, Economic value added, and Shareholder value) to include other forms of value such as employee value, customer value, supplier value, channel partner value, alliance partner value, managerial value, and societal value. Many of these forms of value are not directly measured in monetary terms. -- Wikipedia

Posted by Malcolm Ryder at December 3, 2005 9:18 PM

Trackback Pings

TrackBack URL for this entry:
http://www.malcolmryder.com/cgi-bin/mt-tb.cgi/176

Comments

Post a comment