Personalized URL's - PURLS

A personal landing page fashioned around a specific customer is a great integration of a customer database with online technology. Offers of interest and meaningful incentives can be presented rather than the "one size fits all" approach. According to Andrus (2008, p1), a PURL "creates a one-to-one dialogue with consumers...." She goes on to say that activity tracking can be used to "contiually refine the content or offer to that specific individual."

Andrus quotes David Rosenthal that over 30% of direct mail recipients prefer to respond online. I know that's true for me. One concern that people have about PURL is the publication of PII, personally indentifiable information. Rosenthal recommends using pass codes to protect the Web page in such cases.

References
Andrus, A (09/01/2008). Personalized URLS. Marketing News. Retrieved on August 26, 2009 from EBSCOHOST.

Causal Maps

As discussed in the previous posting, we determine the stability of a causal loop by offsetting plus and minus segments in the loop. A Causal Map is a hierarchy of interacting, interlocking causal loops. A map represents the overall model. An important question for such a model is will it reach equilibrium based on the nature of casual loops within it.

There are several strategies for determining the stability of a map that can be generalized into two categories: 1). Loops within a map are of unequal importance; 2.) Loops are of equal importance.

If loops are of unequal importance, one strategy for predicting equilibrium is to assume the fate of system is determined by most important loop. An issue arises that the judgment of loop importance may be arbitrary. We can address this with knowledge of the process being modeled to argue about the most important inner dynamic.

On the other hand, we can also neutrally look at which loop has the most variables of interest and asssume such a loop is most important. The greater the number of inputs to and outputs from a variable, the greater its importance. The loop with the greatest number of important variables is the most important loop. This is also known as the degree of the system.

In the example of urbanization that was modeled in the previous posting, variables P, M, S all have more than one output, as do variables P, D, B. So P,M,S,D,B are most important variables. We look for the loop that contains greatest number of these: P-M-S-B-D-P is that loop.

Causal Loop P-M-S-B-D-P has an even number of negative signs so it is deviation amplifying. We would conclude the system represented by the entire Causal Map is unstable, using this strategy. It will eventually destroy itself if something is not dramatically changed.

Now let’s assume a different strategy category for P-M-S-B-D-P –that the loops are of equal importance. To predict system stability, count the number of negative loops. If this is an even number, then the system is deviation amplifying, if odd then stable. In the urbanization case, we have one of the four loops negative, P-G-B-D-P. If the relationships are of equal importance, in our example the result is a stable system. The system will ultimately reach an equilibrium.

Another tactic for a strategy that assumes the causal loops in a map are of equal importance is to count the number of negative relationships between variables, making sure to count a relationship more than once if is in multiple loops. Relational algebras can be developed for such calculations for maps of any size.

It’s possible both strategy categories are applicable to a map, albeit at different times. Urbanization, in our example, may be stable for some period of time when all causal relationships are more or less equal in effect on overall system. Imbalances can, however, accumulate over time to change the egalitarian inter-relationships.

A differential speed that cycles are completed, or a growth in the number of times a particular loop is activated across periods of time can result in an aristocratic set of relationships – some holding more power and control over the system. Then we would change from a strategy assuming equal importance of all loops to analyzing the most important loop.

In sum, Causal Maps are a means to portray a complex interdependence so one can better question the situation. They help to impose some order to the domain being analyzed.

Bounded Rationality

Karl Weick (1979, p 20) discusses the concept of bounded rationality, a concept that is applicable to communications and to information systems. Bounded rationality means that all of us have perceptual and information processing limits. We may always intend to act fully rational but usually we act on easy to get to knowledge, use undemanding rules to search for a conclusion, and take shortcuts whenever possible.

This implies that we need to assume that the decision makers in our communications or information systems may use limited rationality. They form attitudes and opinions, or make decisions in terms of familiar facts and abbreviated analyses.

Weick’s discussion of Bounded Rationality extends earlier work done by Simon (1960). Simon (pp 80-84) analyzes the limits of rationality. He finds that behavior is not objectively rational for three reasons:

1. Rationality requires complete knowledge including the anticipated consequences
2. Consequences are future events so impacts can only be imperfectly anticipated
3. Even if all possible alternatives are known, it is unlikely the decision maker would be able to recall all of them in the decision making process

The needed abilities for objective rationality are at odds with the usual reality of fragmented knowledge. Objective rationality also runs counter to the devious consequences of indirect influences in a casual map. Finally, it is not reasonable to assume that all possible alternatives could be considered in a reasonable timeframe, even if they are known.

Simon concludes (p 108) that

“Human rationality operates, then, within the limits of a psychological environment. This environment imposes on the individual as ‘givens’ a selection of factors upon which he must make a decision.”

The implication of this, according to Simon, is that a deliberate control of the psychological environment can manipulate even “rational” choice or decision.

References
Simon, H.A. (1960). Administrative Behavior: A Study of Decision-Making Processes in Administrative Organization. Macmillan.

Weick, Karl (1979). The Social Psychology of Organizing, 2nd Edition. McGraw-Hill.