Decision-making pitfalls for leaders from Chelsea Troy
I recently came across Chelsea Troy’s website, she shares many interesting thoughts there, I recommend checking it out: https://chelseatroy.com/. She is a Senior Staff Engineer at Mozilla and a Computer Science teacher who provides great insights on technical topics, strategy, and leadership.
Her post discussing common decision-making pitfalls for technical leaders was very informative. Pitfall #2, in particular, about treating everything as an optimising metric, particularly stood out:
Pitfall #2: Treating Everything as an Optimizing Metric
What it is: Making a long list of characteristics to consider while making a decision between two or more dependencies, architectures, or strategies, and then forestalling or repeatedly reopening the decision until one option outperforms all the others on all of the characteristics.
I have also fallen into this trap in the past. I remember being part of a team that was incredibly indecisive about platform options. This led to endless debates, a prolonged period without a technology choice, and a lot of uncertainty. We tested countless options, suffering from “death by comparison tables.” And because we could, we would even reopen discussions six months later. Not fun at all!
I like Chelsea’s idea on how to prevent or move away from this situation:
To explain this, I want to introduce you to the idea of optimizing and satisficing metrics.
Optimizing metrics are the ones for which more is always better… Satisficing metrics are the ones for which we have a clear idea of how good is good enough. More of this does not improve our product outcome.
When you are making decisions, narrow down the list of optimizing metrics to as few as possible—zero if you can, one or two realistically. Then, establish the threshold at which your options would satisfice on all of the other metrics.
This simple distinction is very much to the point.
In my own experience, getting some extra support and having central IT make up their mind, along with some help from platform providers, helped us move forward. This aligns with the idea of getting clear on what “good enough” looks like for certain criteria.
You can read Chelsea’s full post here: https://chelseatroy.com/2024/10/16/decision-making-pitfalls-for-technical-leaders/
