Decision modeling together, to save the world

Organizations like the Millennium Project, certain government agencies, companies from the smallest to large transnationals, and the United Nations have recognized that, in a complex world, the risk of decisions that lead to unintended consequences are increasingly great.

In addition, many of these companies are looking to supercharge the effectiveness of their decisions by using advanced technologies like AI and others described in my upcoming book, Link.

Surprisingly, the vast majority of these companies do not consistently use a formal process for decision making.  One senior government official told me that his team simply gives the same budget to each department that they have received in previous years, and that this approach is widespread.  Another senior strategy officer at a large telecom company told me that he receives a sheaf of powerpoint slides every quarter.  After poring through the data, the decision is made without using it in a structured way.  A third friend of mine, who is arguably the country’s leading city sustainability officer, uses a massive spreadsheet that he says he no longer understands.  A fourth friend who is veteran software architect for a large telecom company says that pricing decisions are made informally, again using a massive spreadsheet that is full of errors.  And the list goes on, to dozens of anecdotes I’ve heard over the years in many different industries.

This failure in structured decision making is particularly surprising considering the trillions of dollars at stake, and in some cases human life and health.  You would think that, the more that was at stake, the more rigor is applied to the decision.  Indeed, the opposite seems to be true: as decisions are made at higher levels in organizations, and more is at stake, they become more informal.

There appear to be two primary reasons for this.  First, the complexity of coordinating multiple departments is overwhelming, so in a sort of “analysis paralysis”, senior managers tend to fall back on intuitive, not evidence-based, decision making.  Second, there is a substantial cultural, language, and knowledge gap between data-focused teams and management.  This is not unlike a team that sets out to build a skyscraper, and, lacking a blueprint, the construction crew is unable to build what the customer wants, and the customer is unable to effectively communicate his goals, nor translate those high-level goals into detailed elements like windows, doors, and rebar.

Decision modeling has the following benefits:

  1. To align a team around a common mental model for the situation in which they are operating.  From this point of view, simply drawing the decision model diagram together as  a team can be of tremendous value.
  2. To discover points of greatest leverage in a decision.  What decision, once made, will have the maximum benefit with the least cause? (These are Buckminster Fuller’s Trimtabs).
  3. To discover shared interests and shared solutions in situations that would otherwise appear as conflicted.   Decision modeling has the ability to identify decisions the “holy grail” decisions that maximize the interests of multiple stakeholders.
  4. To avoid unintended consequences.
  5. To act as a “blueprint”/requirements language for the insertion of advanced technologies like AI, ML, optimization, and more into organizations.
  6. To form a basis for continuous improvement, because the decision model diagram is updated over time as more and more is known about the situation.
  7. To determine where will be the greatest value in gathering data.  All data is not alike: some has a much greater impact on a decision outcome than others, and so gathering data in advance of building a decision model can lead to substantial wasted time.  For instance, we may realize through decision modeling that it doesn’t matter if our competitor charges $1 or $10 for a product: we’ll still make the same decision, so refining this value further is not of value.
  8. To determine where will be the greatest value in obtaining additional expertise / facts.  A similar argument to the above one applies here: for instance we may realize through decision modeling that determining the color of our competitor’s product is surprisingly very important. 

By bringing computer technologies like AI, complex systems, and big data, together with human expertise as modeled in a decision model diagram, it is possible to supercharge the above benefits to a level that is only possible when both of these information sources are combined together. This is the “holy grail”: the great promise of the solutions renaissance.

Learn more about decision model diagramming in these posts and in my upcoming book, Link.

Excerpted from Link, in press, Emerald Press, 2019. Sign up here to receive notice of publication.

Lorien Pratt

Pratt has been delivering AI and DI solutions for her clients for over 30 years. These include the Human Genome Project, the Colorado Bureau of Investigation, the US Department of Energy, and the Administrative Office of the US Courts. Formerly a computer science professor, Pratt is a popular international speaker and has given two TEDx talks. Her Quantellia team offers AI, DI, and full-stack software solutions to clients worldwide. Previously a leading technology analyst, Pratt has also authored dozens of academic papers, co-edited the book: Learning to Learn, and co-authored the Decision Engineering Primer. Her next book: Link: How Decision Intelligence makes the Invisible Visible (Emerald Press), is in production. With media appearances such as on TechEmergence and GigaOm, Pratt is also listed on the Women Inventors and Innovator’s Mural. Pratt blogs at www.lorienpratt.com.

You may also like...