It’s a well-known problem nowadays: how can we measure the performance and productivity of individual contributors who do non-routine creative work? The best examples are research and development (R&D) teams, which usually consist of software engineers, designers, scientists, architects, quality experts, product managers, and so on. Such professionals deliver results that are hard to get down to simple numbers. Many authors argue that the very idea of measuring individual performance is toxic and may only lead to negative consequences. We tried to challenge this point of view and did an experiment in our team, which demonstrated that individual performance can indeed be measured, even if people’s work involves creativity, and results are hard to predict. We designed a system of Calibrated Achievement Points (CAPs), which are rewarded to those who deliver visible and tangible results of different kinds. This article explains how CAPs work and summarizes the results of the experiment.
“Competitions are for horses, not artists,” said Béla Bartók, a Hungarian composer about a hundred years ago. Indeed, how can we measure the productivity of someone making a piece of art, say, a painting? By the amount of acrylic put on the canvas per minute? Or maybe we ask painters to compete on the size of the canvas in inches—the bigger the painting, the better the painter!? Obviously, such metrics would be considered not only useless, but harmful. Encouraging artists to compete by such productivity indicators will kill the very idea of art and will most probably discourage most talented artists from participating.
It seems that these days, however, there are people who would disagree with the famous composer. For example, Jacob Eisenberg from University College Dublin and William Forde Thompson from Macquarie University, in their research into how amateur musicians’ creativity changes under stress of competition, demonstrated: competition positively affects creativity. A more recent study by Daniel Gross from Duke University confirmed this finding and extended it with a warning: heavy competition drives us to stop creativity altogether. Anil Rathi in HBR noted, from a business perspective, that “well-managed internal competitions amplify the company’s overall creative ambience.”
Thus, it seems that in the modern world competition and creativity go together. However, the positive effect of competition depends on the rules the team has to play by and the quality of metrics. If the metrics are reasonable and well understood by all participants, the competition drives the creativity.
Of course, measuring the productivity of an artist by the amount of acrylic they put on the canvas or the productivity of a composer by the frequency of pushing the piano keys would be useless and harmful. There are however many examples of good metrics in art, sport, and science, which have stimulated the creativity of artists, athletes, and scientists for centuries: the Nobel Prize is probably the most famous competition every world-class researcher is dreaming of winning, while the Olympic Games and World Championships are what we have for athletes, and the Oscars and Cannes Film Festival are competitions for the best artists in cinema,and so on.
Even though the value of competition may be well understood by a business, a practical application of it in an R&D department may face many obstacles. The main problem would be to answer the question of both employees and the management: “What are the metrics that can objectively measure the performance of our people, who are not athletes or actors?”
After more than a year of experiments we found an answer and built a system of CAPs. We defined a list of 30+ possible achievements any member of our R&D team can make while working within the scope of our research and development. Here are some of them as an example (the “limit” in brackets is how many achievements of this type are possible to count in one year):
- “Major Product Release”: 30 (4) — a software product in a major version is packaged and released to its users;
- “Conference Article Accepted”: 70 — a new research paper submitted to a computer science conference and accepted there for publishing;
- “Technical Report Accepted”: 40 (4) — a short 2-4 page internal report about technical results is submitted to our review board and accepted by it;
- “New Hire”: 20 (4) — a new employee has passed all required interviews and joined the team;
- “GitHub Star”: 1 — most of our software projects are open source, that’s why we reward new stars obtained in GitHub;
- “Lecture or Seminar”: 5 (4) — an educational presentation made inside the department.
The list is available for everybody in our internal Wiki. Each member of our team can decide for themselves which achievements to deliver in order to win in the ranking. The achievements from the list don’t replace the everyday work everybody is doing but are the “extra mile” anyone can walk in order to stand out.
Some achievements are personal, such as a lecture or a conference article, while others may only be created by a team, such as a major product release or new GitHub stars. Individual achievements are attributed directly to the account of an employee, while team achievements are distributed equally among team members: we have seven teams in our department.
Obviously, the levels of competence and qualification are different for different people. Some of them are junior programmers, while others are PhD degree holders with decades of experience. Publishing a computer science article for a junior programmer is a much larger achievement (which is harder to do) than for a seasoned expert. To take this imbalance into account we introduced “weights” for each person, which are used as denominators. For example, if Jeff, a junior programmer, with a weight of 10 publishes an article, he earns 7 points (we divide 70 by 10). At the same time, if Sarah, a PhD with a weight of 25, publishes an article, she earns only 2.8 points (we divide 70 by 25).
We organized our results monitoring system with a “push” principle in mind: everyone is responsible for reporting their own results to our team assistant. We don’t “pull” the information from them, they decide for themselves when and what to report. The team assistant collects the data in a simple Excel spreadsheet and emails it to everybody once a week (there are about 50 people in our team). The document contains the journal of all achievements reported to date, the ranking of every team member, and their weights.
Since the information is publicly available weekly, we encourage every team member to dispute the achievements of others, if they feel like it. We don’t have a special auditor for the results people report. Instead, peer review demonstrates perfect results: when someone reports something that violates our expectations of quality or fairness, almost immediately a concern gets raises. Most disputes, which happen every few months, we resolve by open discussions.
As was noted by David Sarnoff, an American businessman and pioneer of American radio and television, “competition brings out the best in products and the worst in people.” The CAP system we designed helped us not only see the best in product by increasing the productivity and creativity of our people, but also decrease the amount of “worst” which inevitably surfaces when people compete against each other. The CAP system gives us the rules, which are fair and transparent enough to significantly reduce the amount of conflicts and of cheating.
Competitive metrics of people's productivity lead to...
— Yegor Bugayenko (@yegor256) December 12, 2021