Setting New Year's resolutions with decision science
I recently took a decision intelligence course instructed by Cassie Kozyrkov. The course lays the groundwork for making good decisions despite our fallible human nature, including a framework for setting better goals. To retain more of the learnings, I put them into practice by setting my New Year's resolutions. In this post, I will share my notes from the course and use one of my resolutions as the applied example.
Priorities and opportunities
Before applying the goal-setting framework, we need to answer two questions:
What are our priorities?
What opportunities do we have?
The task of discovering our true priorities can feel daunting. The course suggests to start with the priorities of others by asking ourselves:
This thing that others care about, do I care about it too?
If I do, how much?
By reflecting on the personal "non-priorities", we clear the way for more vital things to make the list. Once the list of priorities is populated, we keep the most urgent by comparing each. For me, those were about five pressing priorities for the upcoming year.
Our goals must be rooted in our possibilities to avoid setting ourselves up for a lot of frustration. Consider your current life circumstances that could impact your goals, and take those with you in the next step.
Goals framework
One of the most common mistakes when setting goals is creating them too concrete or overly vague. Kozyrkov pointed out that goals that are too concrete could result in rigidity. On the other hand, vagueness lacks measurability and, thus, accountability.
To mitigate this, we can use a framework that proposes layers of goals with different purposes.
We start with the outcome goal. The outcome goal represents "the win that you are interested in". Its characteristics are: "kind of vague, and not fully in your control". This layer contains the grand vision and the result after all the 365 days of hard work.
Next comes the performance goal. The performance goal is "measurable and mostly under your control". It specifies how you aim to achieve the outcome goal but is not yet concrete enough to pinpoint how you will get there little by little every day.
For that, we have the process goal. The process goal is "measurable and fully under your control". It is realistic instead of aspirational, rooted in the existing opportunities. An important note from the course is to never let the process goal take over the outcome goal and not lose the big picture, the "why". If the process goal stops serving the outcome goal it must be changed.
The principal-agent problem
The principal-agent problem is a concept from economics. It describes a conflict of interest between a business owner and an agent hired to manage that business (more here). The solution for this conflict is "creating constraints on the agent's behaviour". The principal achieves this by creating rules that constrain the agent’s decision-making in favour of their interests.
I know you are rightfully asking yourself: what does this have to do with my goal-setting? Kozyrkov invites us to imagine our long-term selves as the "principal", and short-term selves as the "agent". Our long-term self and short-term self have frequently conflicting incentives. Creating constraints that enhance the short-term self to make decisions serving the long-term goal is paramount to creating optimal conditions and decision-making that benefits the big picture.
Or, as the instructor called it - "pre-hack yourself".
A New Year’s resolution example
Let us apply all of the above.
As the primary caregiver of a preschool child, an ML/computational linguistics student and intermittently an employee, I have spent too little time doing something “just for myself” in the last few years. With my child about to go to school in a few months and my studies in a phase where I only do individual research that will soon become a thesis, more opportunities have opened for “doing things just for me” to become a feasible, dependable option.
I started with naming the outcome goal: doing things just for me. While it is a “win I am interested in”, it isn’t “kind of vague”, but really vague. How do I know I have achieved it, and do I even know what it means? I started to think about what could make it more tangible for me. What feels like it is just for me? One of the things that came to my mind was engaging meaningfully with my hobbies. That felt like the right direction, but yet too vague. I probed until I found something that fit the description (“kind of vague, not fully in my control, a win I am interested in”):
Outcome goal: establish a writing habit.
Next, I embarked on finding out what would signal that I had established a writing habit. Following Kozyrkov’s lead with "measurable and mostly under your control", I quickly realised that regularly finishing some work would be the best indicator for it. So that became my performance goal:
Performance goal: publish/finish a piece of work every month.
For the process goal, I had to take a good, hard look at my opportunities. I’ve tried countless times to have a daily, scheduled writing routine, but life with a young child is inherently unpredictable. Between all the illnesses, kindergarten closures and holidays, work seeps into every crack of free time that opens, especially when work is research. I knew my process goal had to be very flexible, so short that work would not claim it, and not every day.
Process goal: write 20 minutes, 5 times a week.
To set myself up for success (and circumvent my short-term self), my constraints will be more permissive rather than constricting:
Twenty minutes in total, chunks also count.
Writing can be anything: a blog post, a story, a poem or even a LinkedIn post.
Editing and rewriting also count.
…and done!
Conclusion
Happy goal setting! Let me know if you use the framework to set your own goals. Of course, the course itself is a clear recommendation and has much to offer than just this minuscule part I presented here. Thank you for reading, and I wish you a fulfilling upcoming year.
