Tuesday, July 14, 2015

A Quick Rundown on Performance Appraisals and Reviews.

The topic of annual reviews comes up pretty often. There has been a gathering tide of organizations moving away from traditional performance management techniques because they are perceived as an expensive process that does not actually improve performance, and because they seem to actually do damage to a company's talent pool, culture, and thereby to its reputation in the job market.

Deming was the first voice I heard talking about performance reviews. He said of the review “It nourishes short-term performance, annihilates long-term planning, builds fear, demolishes teamwork, nourishes rivalry and politics.” Deming believed in the system, and that the system of work is the largest determinant of quality and productivity.  Deming also called for the elimination of annual individual performance reviews.

People within factory system largely did the best they could under the circumstances and had neither the tools nor the training nor the autonomy to make the work better.  He suggested that the efforts of individuals did not account for most of the problems (he estimated 6% of production problems are because of individuals) and so rather than focus on the people in the process, the process itself needs to be improved. His work was the basis for the Toyota Production System, and for much of what we recognize as Lean process. 

The belief that "good work comes from good workers" is not entirely wrong, according to Deming. Only 94% perhaps.  Deming stated that "a bad system will beat a good person every time."  He was very outspoken on the appeal of appraisals to management, and the damage that they do. 

In matters of performance management, asked "what should we do", Deming would defer to Peter Scholtes.

In our world of software development, the system is not a system of physical motions and motivation, but instead we have a series of technical/intellectual challenges and short timeframes. In that regard, we sometimes resemble inventors or artists or contractors, but our work is of a non-physical nature. As such, our systems have to be optimized for making good decisions, which I have covered elsewhere. 

The need to be productive in this world has spawned Agile methods (though they are sometimes also abused) and the pooling of mental energy such as swarming, pairing, and mob programming. These seem to be leading the charge forward, but they also make it hard to assess the work of individuals. 

Which is why individual performance review in software companies (and similar industries) is unlikely to last, and systems built to support those individual performance reviews will probably have to be dismantled and may not even be replaced. 

Still, many companies believe that performance reviews can motivate employees, recognize top-performers, reduce turnover and attrition, and protect the company legally.  That last argument is pretty strong glue for holding a practice in place. However, in recent years we are finding that it is also not true (enough). 

Enough preliminaries.  Here are some quick links to articles I have selected.  :

CIO Magazine published Esther Derby's classic "Stop Demotivating Me" which is not strictly about performance review, but has influenced most of my thinking on the topic. I suspect that most "performance management" is measuring how people hold up under demotivating circumstances. More is covered (with references) in the Scrum Alliance article "Performance Without Appraisal." Provided that you are compelled to perform annual reviews, Derby also gives advice on avoiding pitfalls and making the best of it.  If you are going to skip performance reviews, she gives advice on what to do instead.

Of course, one would be remiss if one did not include the opinions of the wonderful and thoughtful Bob Sutton, as he asks if reviews do more harm than good.  General opinion: if Bob Sutton writes it, take the time to read it.

Likewise, the illustrious David Rock, along with Josh Davis and Beth Jones, say that we should simply kill our performance ratings.

Samuel Culbert  (UCLA Anderson School Of Management) says that reviews are painful, expensive, and aren't working. He gives ten reasons.

The Harvard Business Review discussed Deloitt's journey to reinvent performance management.

Forbes published "3 Reasons Companies Should Abolish Employ Performance Reviews Now" by David Williams (of Fishbowl) which is more about why NOT to do them, rather than how to supplant them.

Also from Forbes, Josh Bersin asked if it was time to scrap, radically change, or re-engineer the process.  He follows his question with "the new keys to success" to help organizations find a way to work without annual reviews.  His suggestions on self-assessment are good, but I keep thinking about both about the Dunning-Kruger effect and also that people have motivation to assess themselves more highly in order to gain more income. If it were not for the other balancing ideas, self-assessment could be as risky as prior systems.  It also has a focus on individual performance that I think misses the mark (Deming would not approve), but it's a step in the right direction.

In the Wall Street Journal, Samuel Culbert offers another alternative method which still focuses on the contribution of individuals.

In Bloomberg, there is a cautionary tale of the "culture of domination" that grows around reviews.

Adobe documented their elimination of performance reviews.

Suffice it to say that this "kill performance review" discussion is not the work of uneducated, non-managers who don't know any better and who are angry about their review being negative. This is the work of studied, experienced, thoughtful managers and companies who have recognized that the system as-given is not only failing to reach its benefits, but is causing damage in many companies.

From yours truly, here is an  earlier rundown, a bit about rewards and performance, some more on performance bonus, and an old motivation post from 2010.  You might find some advice in there that is of use. If so, let me know.

No comments:

Post a Comment