Signals in R&D

This may seem a strange website to be writing about demagoguery+ but wait for a few paragraphs of political philosophy and the reasoning will become clear.  The demagogue is that smooth talking, slick looking, highly educated professional who will try to convince you that he or she knows what’s best for you. They play on your fears. They play on your sense of fairness. They play on your need for order and predictability. The demagogue is expert in the use of statistics. They have data and statistics backing all their arguments. They are adept at making the statistics sing: there is not a statistic made that cannot twisted to their purposes.  Not surprisingly all the statistics point the your need to elect this demagogue as your leader to save you from some looming disaster.

The patriot, on the other hand, fears the tyranny of statistics. The patriot knows that scientific knowledge (i.e., systematized, statistically-adjusted knowledge) omits the 90% of the contextual understanding needed to make a sound decision. The patriot understands that highly educated individuals belittle the vast amount of circumstantial knowledge possessed by non-educated individuals. The patriot believes in an organization where each individual, only in possession of his or her circumstantial knowledge, can be made to fare far better than a centrally planned organization, by any measure, including fairness. The patriot believes overwhelmingly in individual freedom. Note: I focus on freedom not with the intention of making any political statement, rather because Freedom & Responsibility is one of the selected Behavior Fundamentals+ used in this website.

Organizations lacking solid metrics (i.e., interim progress) are the ideal breeding ground for the demagogue. Demagogues thrive in the informational cesspool that is today’s R&D organizations. While there are leaders that truly know what to do, there are many more who merely think they know what to do. Since there are no solid facts to the contrary the later are frequently convinced they are doing the right thing.

Signals+ are solid metrics. They are the byproduct of innumerable individual transactions – individuals sending signals without even knowing they are doing so. These signals are not subject to corruption, distortion or co-option. They can be subject to speculative bubbles, so to speak; of this we need to be constantly attentive, but for the most part they often provide the only true measure of progress in an organization.

Signals are the best, most reliable means to stop the demagogue. Demagogues hate signals. To take an example from political science, if you raise taxes on capital gains then total revenues to the government drop as many more individuals decide they will hold on to their stocks instead of realizing the capital gain. It’s an unmistakable signal to the government that raising capital gains taxes is a bad idea. Good signals are the bastion of individual freedom – you know how well (or not) your really doing and when you take corrective actions the effects can readily be measured. Freedom, and the signals that support it, require constant vigilance.

Signals are a special type of metrics: they are impervious to measurement distortions because individuals act as though they are unaware they are sending signals, even when they read about them, say, in this forum. Traditional corporate metrics are deliberate and conscious, and often subject to distortion.

Individuals who have not studied micro economics might be surprise to learn that this academic discipline has move far beyond discussions of money or of supply and demand:

If we can agree that the … problem [of Early Stage R&D] is mainly one of rapid adoption to changes in the particular circumstances [during development of the prototype+], it would seem to follow that the ultimate decisions must be left to the people who are familiar with these circumstances, who know directly of the relevant changes and of the resources immediately available to meet them. We cannot expect that this problem will be solved by first communicating all this knowledge to a central board which, after integrating all knowledge, issues its orders. We must solve it by some form of decentralization … but the “man on the spot” cannot decide solely on the basis of his limited but intimate knowledge of the facts … There still remains the problem of communicating to him such further information as he needs to fit his decisions into the overall … system. Adapted from Hayek, Friedrich A. (1945). The Use of Knowledge in Society, H.17

Modern micro economics is all about information, and how we get individuals to reveal their information both before and during a transaction (e.g., a decision based on that coin-of-the-realm called Evidence). The transaction can be monetary, tit for tat or merely a wink and a nod. And the economic discipline has a lot to say about how to make this revelation+ more efficient, with signals at the heart of the answer. 

Individuals reveal their hidden information through signals. For example, sales personnel are keenly aware of signals from their customers. Good salesmen subtly but actively promote signaling by their customers.

Sales Manager:    How did you do with that customer?
Salesman:    I did great. We agreed to get together again next month.
Sales Manger:    Did the sale advance? Did the customer agree to put in some effort on her own part? Is she going to set up a meeting for you with her boss? Is she going to get you the Master Services Agreement? What advanced from your sale?
Salesman:    She seemed to like my presentation and we agreed to meet again next month. I’m still in the door.
Sales Manager:    Well sometimes people are just too polite. If you didn’t get her to commit to doing something that shows some skin-in-the-game then there was no advancement.

The number of sales calls is irrelevant. Only the numbers of advancements+, the signals of the willingness of the client to buy, matter.

What we seek in World Class R&D is in a sense a giant signaling mechanism: individuals signal their belief in the direction of research. If you use Trial by Jury+ as a decision structure then you have signals that the quality of our evidence is improving (or not) captured in the transcripts from the trials. Important additional signals are found in the schedule changes in the court docket. If you use Independent Evaluation+ of R&D projects then clear signals of team productivity can be found in the aggregate of their Consumer Reports+ grades. Many additional signals can be designed into the work activities of R&D that are not found in today's R&D organizations.

Signals are everywhere. They give the outsider access to the internals of the transaction. They allow us to know whether our mechanisms are really improving effective behaviors or not. A wink and a nod signals implicit knowledge+ shared by the team or in the relationship. An advancement as illustrated above allows us to know the level of commitment one individual has for another. One major pharmaceutical company reportedly tracked (surreptitiously) research staff enthusiasm by counting vehicles in the parking lot after 6 pm.

For industrial R&D I advocate two promising signaling mechanisms: one monetary (Employee Stock Option Plans) and one subjective (Implicit Trust+). The beauty of a monetary transaction (as when you use a cash register) is the economic signal is objective and easily captured and aggregated. Most interpersonal interactions, however, are subjective and present challenges in both the capture and the interpretation of the underlying signals. The ESOP+ signal captures the direction of rational (self-interest) behaviors. The Trust signal captures the direction of emotional (self-effacing) behaviors.

Home Page April 2010