Besides the fact that this article describes indeed an interesting practice of hiring process improvement, I also see it as a great example of how to formalize the feeling of needed changes for any process, functionality, or even Objective from the OKRs list.

That's how I see the steps:

1. formulate the problem. It should not be the final, well-stated problem, just write down your concern with some background info;

2. create a Request-for-comments doc to collect suggestions from various stakeholders about how to solve your concern;

3. hold some sessions to review the comments. At this moment, it's highly possible that your understanding of the problem will expand, and suggested ideas will transform into new ones. It's time to define your strategy for problem resolution;

4. for each group of ideas, define metrics to measure the progress. Select the north star metric and its value;

5. experiment!

(1) Select stakeholders that are the most interested in the change and do not afraid of being early adapters; (2) hold impact mapping sessions and bet on some improvements to try them in the first place. Define the fundamental part of your changes; (3) validate the progress by the metrics regularly; (4) create a pipeline to visualize your progress;

1. analyze the results of the experiment. Adjust the changes made before if needed.

2. In case of positive results, spare the time for less significant changes and continue to track the progress.