I was crossing a traffic signal the other day, and there was a car waiting to turn left after managing both myself and the incoming traffic. The driver was visibly irritated as the signal turned red, meaning they would have to wait for the signal to turn green and repeat this whole exercise.
The first thought that came to my mind was how useless left turns are and how could UPS have come up with a policy to never take left turns and whether today’s modern day Product Mangers can learn something from this decision making process?
What is so bad about turning left?
According to Tom Vanderbilt, author of the book “Traffic: Why we drive the way we do”, turning left can be dangerous as the vehicle has to turn against the flow of incoming traffic. Of all the crashes that occur while turning or crossing an intersection, 61% of those occur while turning left. Moreover, left turns are also three times more likely to kill pedestrians, than right ones! 
According to Jack Levis, UPS Senior Director of Process Management, a left-hand turn is less fuel efficient as the car is idling longer. It’s not that the trucks don’t always turn left, but avoid making left turns unless absolutely required. And as shown in the image above, UPS does save 10 million gallons of fuel every year and that means that turning right is good for the environment as well.
How did UPS come to this conclusion?
The more interesting question here is not the fact that left turns are more efficient, but the planning and decision making that went behind it to make such a radical move possible. The founder Jim Casey, had this philosophy of “constructive dissatisfaction”, meaning that things can always be a little better and the company would try different ways to optimize their systems and processes such as introduction of automatic car washes, not hiring family members, the brown uniform etc. 
UPS experienced significant growth in the 1960s and had to rely on operational efficiency to control dispatch. As the modern digital technology was not available then, UPS had to rely on historical information to plan routes. The entire service area of a package center was divided into smaller areas called loops, and Industrial Engineers would arrange street segments into a loop’s AM group and a PM group.  An informal no-left turn policy is said to have evolved from this loop dispatch concept, though I could not find any direct evidence to prove so.
UPS knew that they had to make use of technology to scale their operations further, and hence decided to give UPS drivers a handheld device called the DIAD (Delivery Information Acquisition Device). The device not only saved paperwork, but the data generated from this device provided UPS with useful data to further optimize their operations.  This then further led to the evolution of the in-house routing system called ORION that adds a 20 second penalty for left turns, to generate the most effective delivery route.
What can we learn from it?
Modern day Product Managers have to make such decisions everyday on how should they pivot their products to achieve their metrics and OKRs. Getting ubiquitous data is not the problem today and some people advocate towards going by their intuition first and then using data to validate their assumptions, while others advocate leaning more towards data first and then finding insights from it.
What I learn from the UPS story is that even though we might be meeting our numbers, the first step comes from the willingness to optimize and do better. Secondly, just looking at data is going to give a very aggregated and macro level picture of what’s happening. Though useful, actual insights would unveil faster by combining data with localized knowledge so that we know why the data is behaving the way the way it is.
The way UPS pivoted to relying on technology for its operations highlights a very important phenomenon to take a step back and develop long-term thinking. Sometimes when we obsess about a metric, we treat it as an end-goal and fail to look at the big picture to see whether our solutions are actually treating the disease or just the symptoms in isolation.
Do let me know what you think and if you liked reading this post, here’s a few that I can recommend:
Addressing the disease, not the symptoms
I don’t think there’s a literal definition of the term disease, but the closest I could find is a harmful deviation…
Will there be a COVID vaccine black market?
If limited supply, and high demand is a good indicator for black markets, there is no more suitable candidate for it…
Incentives for the long game
Circling back to Charlie Munger’s ideas about incentives, and thinking whether there’s an incentive system to keep…