It started with a problem.
The first electric traffic signal was installed at the corner of Euclid Avenue and East 105th Street in Cleveland, Ohio on August 5, 1914. The device had red and green lights operated manually by a police officer in a booth, with a buzzer sounding when lights changed. Garrett Morgan, an African American inventor and businessman in Cleveland, developed a three-position traffic signal that he patented in 1923. Morgan's design added a neutral position — a yellow warning light — between stop and go, giving drivers more time to respond to changes. He sold the patent to General Electric. The three-position signal became the basis for the modern traffic light.
Historical sources below ↓Then something changed.
The traffic light removed the need for a police officer at every busy intersection. A rule — encoded in the timing of colored lights — replaced a human judgment about when to allow cross-traffic. This made the system scalable: the same rule could operate simultaneously at many intersections without additional labor. It also introduced a tension that has accompanied automation ever since: the rule is consistent and predictable, but it does not respond to unusual situations. A police officer could wave through an ambulance; early traffic lights could not. Modern systems address this with sensors and override mechanisms — layers added to handle the cases the base rule missed.
What a builder can learn.
Our interpretation: the traffic light shows that replacing human judgment with a rule works well when the situation is predictable and the cost of the rule's failures is acceptable. When those conditions do not hold, the rule needs exceptions — and exceptions need a mechanism. For a business builder automating a process, the questions are the same: under what conditions will the rule fail, how often will those conditions occur, and what is the cost of each failure compared to the cost of building an exception mechanism?
A door still open.
A hypothesis worth testing: if a small business identified one decision its team makes repeatedly by judgment — assigning a task, approving a request, routing a customer — and replaced it with a simple documented rule for one month, how often would the rule produce the same outcome the human would have chosen, and how often would it produce a worse one?
Which recurring decision in your business could be handled by a rule — and which conditions would cause that rule to fail in a way that matters?
Follow the evidence.
History is sourced below. Business lessons are editorial interpretations; future possibilities are questions to test, not predictions.
Prepared from primary-source histories. First edition: 25 Sept 2026.