Personality assessments

I’ve been really interested in personality assessments lately. The Myers-Briggs (MBTI) test is of course the classic, but I came across the “Team Dimensions Profile” through Ray Dalio’s excellent book Principles and find it to be the most interesting and useful. (I’ve also been meaning to take the Workplace Personality Inventory, but cannot seem to get access to it without impersonating an HR professional — any ideas?) I have found two primary uses of these types of personality assessments. First, they shed light on the types of tasks and roles within a team that I am most naturally suited for. This is useful for thinking about where I can have the most impact and be the most successful. Second, they help me understand other people.

Pretty much anything worth doing requires a team of people with diverse abilities, skills, backgrounds, and personalities. But the aspects that make a high-quality, diverse team potentially so powerful — different perspectives, different working styles, thoughtful disagreement, etc — are precisely the things that can make it difficult to be a part of one. Understanding how one’s own thought processes and natural tendencies differ from those of your teammates increases mutual respect and the quality of the work that can be done together.

The Team Dimensions Profile

The Team Dimensions Profile evaluates based on two independent metrics: (i) whether you tend to think in terms of possibilities or realities, and (ii) whether you tend to focus on analysis or interaction. The result is a point in a disc, with the angle representing where you fall according to these two axes and the distance to the origin representing how strongly. For example, when I took the test, this is what I got1:

The result is a division into one of five rough categories: Creator, Advancer, Refiner, Executor, and Flexer. I’ll defer to the detailed result report for the precise descriptions of the categories. An oversimplified version (also taken from the team dimensions profile result report) is as follows:

  • Creator: generates ideas
  • Advancer: communicates ideas
  • Refiner: challenges ideas
  • Executor: implements ideas
  • Flexer: steps in to fill in the gaps

For Teams

One reason I like this classification system is because it parallels the actual steps involved in doing anything nontrivial. And since anything nontrivial that is worth doing will likely require a team to do it, it makes sense to understand where you and your teammates fit into the process. It also makes sense to understand if your team is missing a person with one of these natural tendencies.

Roughly speaking, the steps involved in solving a problem look something like this:

  1. Creation: start with a broad discussion of what is possible — without worrying yet about how practical it is or how it will be done. Then narrow down the set of possible solutions with preliminary considerations of feasibility to arrive at a few options.
  2. Refinement: critically examine the proposed solutions and try to poke holes in them. Decide on the best solution.
  3. Execution: design a strategy for implementing the proposed solution, discretize it into a steps, and then carry them out.

Each of these steps requires different types of thinking. Different people will take different amounts of satisfaction and enjoyment from each step. In a highly functioning team there will be a natural handoff process, in which work flows from one person to the next according to their preferred thinking style. (This application of the team dimensions profile is a modified version of the “‘Z’ Process” discussed in the full result report.)2

But there is another way of looking at the implications of the team dimensions profile classification which is equally as powerful. Instead of thinking of people being best matched for the different stages of the creative process depending on their tendencies, it is worth thinking of the different perspectives different types of people will bring at any one of the different stages.

For example, during the creation step creators may suggest solutions that are unconventional and innovative, but perhaps also irrelevant because they ignore practical realities. The executors on the opposite end of the possibilities/realities spectrum may suggest solutions that more directly and efficiently solve the problem. During the refinement step, refiners may notice quantitative inconsistencies in a proposed solution, while advancers may be concerned with inability to inspire enthusiasm from those who will carry it out. And during the execution step, creators and executors will argue over the best strategy for implementation, while refiners and advancers may debate whether the metrics of success should be based on quantitative or qualitative evaluations. Next time you find yourself wondering why on earth your colleague made a suggestion or asked a question that struck you as odd or just plain silly, consider that perhaps you are simply on opposite sides of the team dimensions profile classification. Then be grateful for them, because they’re filling your gaps!

Then there is the Flexer category. I don’t have much to say about this, other than to wonder how the outcomes would compare for a hypothetical team A composed of four flexers and team B composed of four people who are near the edge of the disc, but who span the entire range of personality types.

For Individuals

I will provide an example of how the team dimensions profile classification has helped me understand my own workflow and identify holes, regardless of the role I am playing on a team or if I am working independently. As someone with a strong creator tendency, I get extremely excited when thinking about new ideas and finding abstract connections between them. But I also recognize the value of producing results. In graduate school I learned through trial and error that if I didn’t execute a new idea almost immediately (e.g. within a few days), the next idea would come along and the first would end up forgotten. As a result I subconsciously developed a workflow in which I jumped almost immediately from the creation stage to the execution stage. I then noticed that I would waste a lot of time executing ideas that didn’t work out, that in hind sight could have easily been shot down with a few more minutes of refinement. I now try consciously to slow down between creation and execution to spend more time refining. This is something that I am still working on. The conceptual framework and vocabulary of the team dimensions profile helps me do this more consciously.

Conclusion

I have been fortunate enough to be a part of a diverse team of talented people for the past year and a half at Cuberg. This experience has given me both the highs of seeing what such a team can accomplish and the lows of seeing how interpersonal conflict can stymie progress. Before discovering the personality assessments, I had seen the forces at play but did not have the language to analyze them.

Notes

  1. Note added in 2025: I took this test in 2018. I think if I were to take this again I would get something much closer to the center. Scores on tests like these are shaped by your experiences. While I still enjoy ideation, my experience as a startup founder over the past couple of years has significantly pushed me towards the execution direction. 

  2. It is worth emphasizing that this should not be taken as a limiting classification. One may be able to do a fantastic job in a step in the process that is not their natural tendency. They just won’t be able to do it long before becoming dissatisfied and start seeking something else. A related point is that of course different people will have different abilities, regardless of their team dimensions profile classification. An advancer with an extremely high intelligence will likely outperform an average-intelligence refiner even in a refinement task. 




Enjoy Reading This Article?

Here are some more articles you might like to read next:

  • Capability-based evals for chain-of-thought monitorability
  • Early ideas for mechanistically investigating emergent misalignment
  • Circuit tracing for chain-of-thought
  • Cross-layer transcoders in a battery foundation model
  • A foundation model for battery data