Re: MBO, Perf Reviews, Incent Pay & Quality

GAWNE, SEAN (gawnesm@songs.sce.com)
Tue, 20 Dec 94 22:47:13 PST

This is a long reply, delete now if you like.

Rick Karash asked what an ideal performance appraisal system looks like.

I have always had difficulty with performance reviews. As a participant on
both ends I've always felt there was a problem with trying to distill so
much about a person into a simple report.

The best use of performance appraisal I've seen is when supervisors (those
who care about their people) write more than one performance review during
the year with only one being "official". The others are no more than a
formalized feedback/coaching tool, between a worker and a supervisor or
vice-versa, and generally go nowhere but to the two people involved.

In many union jobs, pay is still not directly linked to performance. Just
as in the military, there are legitimate reasons for this practice.
Ideally, it allows the team to carry the weight for an individual who is
weaker. In practice this may cause tension unless the team is given some
say about hiring, firing, discipline, and so forth.

Performance reviews also seems to be more of a problem today because of the
"dejobbing" phenomenon. When there is little similarity between what one
employee and the next employee do each day, it becomes very difficult to
explain the rationale for paying them a certain wage.

But like most practices which seem so absurd to those of us who don't
remember how they got started, there are good reasons why both employer and
employee may want to avoid a "pay for performance" system.
- when employees are individually ranked and paid accordingly, this has two
undesired effects (probably many more, but two that I hate):
1) the employees are pitted against one another in competition for a
(usually
shrinking) piece of the pie. This is typically in an atmosphere of
increasing need for teamwork, so it's inherently counter=productive. Yet
even in the face of this obvious hypocrisy, many companies persist in this
practice.
2) those employees with good argumentative skills tend to outperform their
counterparts at appraisal time, because of their desire/tendency/ability to
go face to face with their appraiser. Good, hard-working, but modest or
meek employees are run roughshod over. Again, we reward negative behaviors
for the typical workplace. Self-centered, greedy, argumentative types are
not the best for many jobs, but tend to do quite well in performance
reviews.
- this process creates a great deal of stress for both parties involved.
- the process consumes a great deal of time and effort on management's
part. Often the design of the process means that it has little to do with
meeting business goals, so this is time wasted. Managers and supervisors
know this but have no choice but to comply.
- employees in many jobs have little idea of the true value of the services
they provide, and are ill-equipped to discuss this.

Well, the more I think about what's wrong with this, the more I see that
there are deep problems which need to be addressed, and that the
performance review is not necessarily the problem.

Why do we do performance reviews? To periodically stop, reflect on what
we're doing, how well we're doing it, and how we can do better in the
future. It's healthy for a business, and healthy for the individuals who
are part of that business.

How can we make a system that would do that well? 1) It should involve
several types of review:
- the individual's self-assessment
- assessment by peers
- assessment by supervisors, if applicable
- assessment by subordinates, if applicable
- assessment by customers, if applicable
- assessment by others, if applicable (members of other teams, support
staff, whatever, there are so many structures no scheme can meet all
needs.)
2) It should address business needs
- are goals met
- are deadlines met
- are budgets met
- are customers satisfied
3) All parties should be prepared with essential knowledge of issues:
- if pay is dependent on this, all involved in discussions of pay have
an obligation to learn the value of the service provided. Becoming more
aware of the value of services provided is beneficial to both employer and
employee. This should include at least a comparison to typical market value
for this service and an attempt to quantify the actual value to the company
of the services provided.
- goals, deadlines, budgets, quality criteria, etc ought to be objective
and there ought to be consensus of parties involved. Consensus meaning
informed consent, agreement to support, not necessarily unquestioning
loyalty.
- company resources, plans for the future, and reasons for decisions that
affect the employees need to be openly discussed with employees. 4) All
efforts should be made to keep the system as simple as possible. 5) In an
empowered workplace that wishes to focus on the team, not a traditional
hierarchy of supervisors and workers, the TEAM should largely decide on
fair compensation for the team members. Negotiations about pay would be a
group effort. This reduces the importance of self-promotion skills and
helps promote teamwork. It also defuses many individualistic
behaviors, since the team is the ultimate arbiter and would be expected to
punish non-team behaviors.
6) There is no reason to do this at the same time every year, in fact doing
so usually dilutes the effectiveness because it is typically done at the end
of the year, when many other important tasks are pressing, and everyone is
done at the same time, so it's unlikely a supervisor will be able to give
each review the time it deserves. A better idea would be to try and meet
with each employee a certain number of times a year -- this too will
probably vary from one employee to the next, some people like a lot of
feedback and some don't. Schedule these reviews over the entire course of
the year, a regular activity. This is after all management in practice, the
manager coaching the employees and monitoring progress toward business
goals. It ought to be ongoing at all times, not just once a year.

As we move toward a more project and task-oriented workplace world, and
clearly defined jobs seem to be fading into the past, this seems to have to
become a more important and relevant activity. Much as I might wish it
would go away, it seems to become ever more important.

Some people strongly object to having their peers have any control over
their pay, or having any responsibility to measure their peers'
performance. This may be resistance to change, or it may be a valid
concern. No system works in every situation. I just think that if
"empowerment" is the valid way to do the work, it should also be valid for
this function.

Also it gets a bit difficult in some organization to determine just who
your peers are, or what business unit you belong to. That goes hand in hand
with the dejobbing thing, and makes pay for projects or output more
sensible than regular wages. However both employers and employees are
conditioned to the security of long-term contracts and will accept a good
deal of losses to maintain the long term relation.

I would like to close this by describing what my own company is doing this
year; we are in the midst of many changes. We've come up with a new
performance based reward system, where employees have the chance to share
in profits if the entire company and their business unit exceed certain
goals in 1995. This is far different from past systems. A milder version
adopted last year has been abandoned. Now here's the wierd part: 1994
performance was based on that other plan, which was dropped. We will still
go through the exercise of performance review for 1994 and performance
planning for 1995, but all wages are frozen. So it really doesn't matter
how well you met the goals for 1994. We will just go through the motions.

Suffice it to say I will not spend too much time on the exercise. As you
can tell by the wandering nature of this message I am more than a bit
confused right now, but I hope that someone out there can make something
out of all this. Sorry if you read all this way and feel cheated. I am
feeling a bit cheated myself, actually.

Sean Gawne,
gawnesm@songs.sce.com

"Everything should be made as simple as possible, but not simpler."
-- Albert Einstein