Ashwin's Weblog

Why Performance appraisals don’t work

  • Posted: March 5, 2010
  • |
  • Author: Ashwin Prabhu
  • |
  • Filed under: General
  • |
  • Tags: No tags set for this entry.

March is that time of the year when the usually dormant HR departments at software companies work at their peak, preparing the stage for the next season of Employee Appraisals. And coincidentally, so do the software engineers, forwarding emails dissing performance reviews and grumbling about their managers at impromptu cafeteria huddles.

The most stressful time of the year in the two previous companies that I worked in, discounting any major product release date, was the performance review period. I bet the sight is similar across companies, year after year. Nothing has changed and it seems like nothing is going to change. I have been part of this elaborate drill many times before, and I have also been a witness to the “silly season” that follows.

Both companies that I’ve worked in followed a somewhat similar numerical rating approach. While the former trusted the manager to do all the excel sheet based grading on her own, the later rolled-out an ugly Oracle ERP suite to collect peer reviews in the hope of garnering a balanced feedback. Coincidently though, both the systems worked exactly the same way. The process began with somebody in the HR department creating a questionnaire listing various non-scalar qualities like “Communication skills”, “Lateral Thinking” and a set of numbers (0-5) to choose for each quality, 5 being the highest score. The managers had to rate their reportees against every quality. It didn’t take a genius to realize that the only possible values were 3 and 4, anything less meant trouble. The managers at my first place figured out a creative solution to the problem – fractions. It was not unusual to be rated 3.62 for something as subjective as “Works well with others in the team”! The ERP though constrained the options to whole numbers only. Finally the exercise culminated with the system spitting out a weighted average between 2 and 5, in fractions.

Systems aside, the experience is traumatizing for the first timers, and frustrating as always to others. So why are performance reviews made out to be such a stressful exercise?

Two agendas

When was the last time you bargained hard for a lower price and succeeded, only to find out later that someone was lucky enough to buy the same thing elsewhere at a price lower than what you had paid? Remember that feeling? That’s exactly how the participants feel as they come out of a review meeting. The stakes are high when the reviewee sincerely believes the rating will have a bearing on her future pay.

The manager wants to discuss things such as skill upgradation, team relationships, opportunities for improvements, while the reportee wants to talk about training requirements, career growth and day to day issues impending performance, mundane as it might seem to the manager. The manager feels that her judgment and standards of excellence are better than that of her reportee. For the reportee, it’s like playing against a moving goal post, no fun. The differing agenda will put the two at odds with each other. At best, the discussion will drag for an hour and will accomplish nothing.

Sometimes though, the discussion will turn into an argument, with the reportee concluding the manager to be a living version of the pointy haired boss and the manager subconsciously classifying the reportee as yet another case of “defensive and resistive to suggestions”. More likely this tension will visibly fill over into their daily relationships.

One size fits all

In the real world you never use the same parameters while comparing a Chihuahua with a Labrador. Why software companies keep using the same list: “Commitment”, “Team Player”, “Problem Solving Skills” yada yada… for everyone puzzles me. For the sake of brevity, some companies use a weighed scale while calculating the performance score. That way they can ensure everybody gets rated differently based on the seniority band they fall under. Traits like “analytical thinking”, “problem solving” and “programming skills” weigh more for an entry-level developer. The scale starts tilting more towards soft skills like “communication skills”, “team player” for a mid-level developer, technical skills become less important. No wonder, this has nourished a culture where developers no longer want to code after the initial 5 years. They just want to manage people and steer projects by excel sheets. When was the last time you bumped into a dentist who refused to perform filling jobs, because he has been doing it for the past 5 years?

Performance reviews are stressful for other reasons as well. One of my friends always got lower than expected reviews. He used to spend his personal time tinkering with the latest libraries or in figuring out potentially better solutions to solve the daily issues. He liked to pass of his new found knowledge to his team mates, and their approval massaged his geek pride. His unique abilities did not show up in the list and the manager took his talent as a distraction and mismanagement of his and the team’s time and rated negative. This broke his spirits, and he stopped sharing his tinkerings with his team. Everybody lost. There can be many examples such as these – a developer who is highly productive but rated negative since he refuses to do extra time beyond a certain limit, a tester who has the ‘annoying habit’ of identifying usability issues. The list is endless.

Objective vs Subjective

Invariably reviews are political and subjective. The notion of objectivity is absurd. All things equal, if you are in good terms with your manager you get rated favorably. I have often heard of people getting very different ratings from their new managers within span of months. Despite the best intentions, the system cannot truly eliminate the subjective nature of reviews.


Negative reviews hurt morale, affect productivity, while positive reviews if not channeled properly can have a negative effect in the long run by breeding complacency. Good team bonding is utmost essential to achieve success; performance reviews can potentially undermine this bond, specially the one-on-one relationship between the manager and the reportees.

While it is not right to rate everyone within a team equally, any discrimination in rating, however fair it might seem, will lead to depressed morale and ultimately a few resignations.

Edwards Deming in his book ‘The New Economics’ describes performance rating as:

The idea of a merit rating is alluring. The sound of the words captivates the imagination: pay for what you get; get what you pay for; motivate people to do their best, for their own good. The effect is exactly the opposite of what the words promise.

So what’s your take on the subject?

If you are unsure how you feel about ratings take this short quiz. I scored 22, which means I am confused.

6 people have left comments

Andrew - Gravatar

Andrew said:

I think reviews become more effective when it is viewed as a mechanism to identify strengths, and work on improving them further. Too many managers concentrate on weaknesses alone.

Posted on: March 22, 2010 at 9:31 AMQuote this Comment
Mark - Gravatar

Mark said:

I always began my appraisals with this line “How can I do better in my work?”. This ensured that the time spent during the 1-2-1 review process was productive for me.

Posted on: March 25, 2010 at 12:49 PMQuote this Comment
Ashwin Prabhu - Gravatar

Ashwin Prabhu said:

@Andrew True, I have seen both the types. I have been lucky in the past to have worked with a manager who mentored rather than reviewed performance.

@Mark Thanks for the tip!

Posted on: March 25, 2010 at 3:26 PMQuote this Comment
Ramesh - Gravatar

Ramesh said:

Good article. thank you

Posted on: April 3, 2010 at 12:49 AMQuote this Comment
Louis Vuitton - Gravatar

Louis Vuitton said:

I do not believe I have seen this described in such a way before. You really have cleared this up for me. Thank you!

Posted on: April 8, 2010 at 9:40 AMQuote this Comment
OldTimer - Gravatar

OldTimer said:

Ive worked at software companies both big and small over the last 12 years. And 1 thing I find almost universal about the review process is that it is generally an exercise in reverse engineering. What I mean is that a given dept is told it has x amount of dollars,budgets are created etc..the reviews are then made to have annual increases fall in line with budgets.
After spending a few years listening to managers give both praise and criticism where they weren’t warranted and show a clear and disturbing lack of knowledge as to what my actual job/accomplishments were I began the practice of opting out of these things whenever possible. And if opting out is not a possibility, I try to waste as little time as possible on the process as the input is usually ignored and the difference between a 3 percent raise and a 4 percent raise hardly matters…

Posted on: April 26, 2010 at 8:42 PMQuote this Comment

Leave a Comment-

Comment Guidelines: Basic XHTML is allowed (a href, strong, em, code). All line breaks and paragraphs are automatically generated. Off-topic or inappropriate comments will be edited or deleted. Email addresses will never be published. Keep it PG-13 people!

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <em> <i> <strike> <strong>

All fields marked with "*" are required.