Decision Log

The practice of formal decision logging has roots in military after-action reviews and engineering change control processes that date back decades. In the military context, the U.S. Army formalized the After Action Review (AAR) process in the 1970s and 1980s, which included systematic documentation of what was decided, why, and what happened as a result. In software engineering, Architecture Decision Records (ADRs), a specific form of decision log, were popularized by Michael Nygard in a 2011 blog post that proposed a lightweight template for documenting significant technical decisions. The concept has since expanded to strategic management as organizations recognized the value of institutional memory for non-technical decisions as well.

Strategically, decision logs address several critical organizational challenges. First, they combat hindsight bias, the tendency to believe after the fact that an outcome was predictable. By recording the information available and reasoning applied at the time of the decision, a decision log allows organizations to evaluate whether a decision was sound given what was known, rather than judging it solely by its outcome. This distinction, between decision quality and outcome quality, is fundamental to good organizational learning. A good decision can produce a bad outcome due to factors outside anyone's control, and a bad decision can produce a good outcome due to luck. Without a decision log, organizations systematically punish unlucky good decisions and reward lucky bad ones.

Amazon Web Services provides a notable example of decision logging in practice. AWS teams maintain architectural decision records that document not just what was decided but explicitly what alternatives were rejected and why. This practice has proven invaluable as teams grow and original decision-makers rotate to new roles. New team members can understand the reasoning behind existing systems without having to re-derive the logic or repeat past mistakes. Jeff Bezos has emphasized the importance of distinguishing between reversible decisions (which should be made quickly) and irreversible decisions (which deserve careful deliberation), and decision logs naturally support this distinction by varying the depth of documentation based on decision significance.

Bridgewater Associates, the hedge fund founded by Ray Dalio, has taken decision logging to an extreme with its culture of radical transparency. The firm records meetings and decision-making discussions extensively, creating a searchable archive that allows anyone to understand why investment and management decisions were made. While Bridgewater's approach is more intensive than most organizations would adopt, it illustrates the principle that systematic decision documentation creates compounding organizational learning over time.

To implement an effective decision log, organizations should establish a consistent template that captures the decision statement, date, decision-makers involved, context and constraints, options considered, rationale for the chosen option, expected outcomes and success criteria, and a planned review date. The key is making the process lightweight enough that people actually use it while capturing enough information to be useful months or years later. Digital tools like Confluence, Notion, or purpose-built decision management platforms can make logging frictionless. The most important field in any decision log entry is not the decision itself but the rationale: the why behind the what.