Cognitive Bias in Enterprise Knowledge Selection
In this chapter we will look into the reason that organizations are not adopting enterprise knowledge graphs due to logical errors in judgement. We call these persistent patterns of error Cognitive Bias. When we combine this knowledge with our understanding of the Technology adoption life cycle and Windows of Opportunity we can start to make create a predictive model of when and organization might be ready to adopt enterprise knowledge graph technology.
Being able to predict when an organization is ready to make the transition from a relational dominated world to a graph dominated world does not imply that everyone who is ready to change will successfully make the transition. Sometimes random events occur that block our evolutionary progression.
When I first started to focus on solution architecture consulting, I was frequently hired by organizations that wanted to bring in an objective external consultant to help them evaluate options for a specific business project. Although our book, Making Sense of NoSQL had ample information on how to do this objective analysis, many companies still wanted an experienced outside person to oversee this processes.
Although many of these projects went well, I was often disheartened when organizations didn't make the appropriate choices. I reasoned that they were making political choices, not rational choices based on evidence, and tried to wash my hands of their choices and I moved on to the next project.
Then in 2013, I attended a conference sponsored by the people that developed and used the ATAM process originally developed at CMU's Software Engineering Institute. This was the CMU Saturn Conference that focused on researchers trying to understand the architecture analysis process. One of the speakers was the incredibly insightful Mary Poppendieck. Mary's presentation was all about how cognitive bias has a strong influence on how organizations select any given technology. I was thrilled to finally have a precise taxonomy of the reasons that politics drove organizational decision making.
You can see the slides of my presentation on the application of ATAM to database selection here. You can see that many of the concepts in this book are present in this presentation.
Since Mary's talk in 2013 I have carefully documented many of the bias I have seen in organization decision making. Here are some of them:
- Anchoring bias
- Availability bias a.k.a. memory bias, familiarity heuristic
- Bandwagon effect
- Confirmation bias a.k.a. Fiter bubble
- Halo effect
- Hindsight bias
- Illusory superiority bias
- Framing effect
- Narrative-bias
- Representativeness heuristic
- Silver bullet
- Status_quo_bias
- Sunk cost a.k.a. Gamblers fallacy