[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference 7.286::digital

Title:The Digital way of working
Moderator:QUARK::LIONELON
Created:Fri Feb 14 1986
Last Modified:Fri Jun 06 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:5321
Total number of notes:139771

1233.0. "Let's change the metrics!" by THRILL::WILS (Interprocess Message Queueing P.O.) Tue Oct 16 1990 16:18

      A seasoned manager, whom I respect very much once told me: "there are
plenty of ideas in this company, but there are too few people that can get
them done". I believe that the reason people can't get things done is
because the metrics prevent them from doing so. I would like to show that
the current way employees are measured drives them to increase cost and
reduce productivity. In turn I will present a new set of metrics that will
drive people to decrease cost and increase productivity. To do so I will
distinguish between managers and individual contributors. 

  The first question any manager is asked when s/he goes for an interview
is: "how many people have worked for you?".  There are other major metrics
that a manager gets evaluated by, but the number of people that have worked
for a manager usually plays a very major role in determining if s/he is
qualified or not. Therefore to be marketable a manager is driven to
increase and not decrease the number of people that work for him/her. 

  The first question any individual contributor is asked when s/he goes for
an interview is: "what is your primary specialty?". A person's specialty is
usually a show stopper in determining if s/he is eligble for a job or not.
There are other metrics, but these are usually of less importance.
Therefore to stay marketable an individual contributor strives to
specialize as an individual and not to diversify as a teamplayer. It drives
him/her to stick only to his/her own direct responsibilities and not to
help someone else out. The result for the company is that resources are not
utilized well. Someone who is stuck and needs to be helped will not be
helped and someone who would do better to teach than do it alone will not
do so. 
  
  No one evaluates a group. And everyone is saying that we need teamwork.
However if a group functions well than the only one that reaps the real
benefits is the group's manager, because s/he will get a good evaluation.
The group is still stuck with the fact that the manager has to spread pay
raises over the group members. Those that do well compared to their
colleagues get more, others get less. The success of the group as a whole
does not benefit them. Even though it is the teamwork among the individual
contributors that results in success. The final result is that people are
driven to stick to their specialties, to compete and not to work together. 

  To break the cycle we must change the metrics. We must choose metrics
that fight the natural tendency of managers to increase the size of their
organization. We must choose metrics that fight the natural tendency of
individual contributors to specialize as individuals and not to work as
team members. 

  My first proposal is to reward groups. A group's manager's manager now
gives a rating to a group. If the group does well it gets a 1, less well a
2 and so forth. If the group gets a 1 rating, then the group's manager will
be allowed to pass out more 1's & 2's to the group members than if the
group got a 2 or a 3. If the group gets a 3 rating then only very few 1's
and 2's may be passed out. This metric will drive individual contributors
to work together and to be good at their own jobs at the same time. 

  My second proposal is to set a goal for each organization on the
following key metrics: output, cost, quality and productivity. This must
start at the very top. Each organization must specify how they measure
output, cost and quality. If one organization's output is the others input,
then the two must agree. For productivity there is only one metric:
output/# of people. Half a person counts as a whole. Once a year the top
must set goals. These have to be reasonable: 10% increase in output at 10%
decrease in cost, at 10% increase in cost and 10% increase in productivity
or something like that. The organizations that make up the top must now
divide these goals among themselves such that the top goal is met. Then
their sub-organizations must do the same, and so forth. During the course
of the year these goals can be changed as circumstances require. 

  A key component of these goals is productivity. This must be measured in
terms of total headcount. Therefore a manager must decrease headcount to
increase productivity if output doesn't rise. To make this feasible the
company must continue to guarantee employment except under extreme
circumstances. Then those employees that are released from a group can be
placed into a TMP like program. They will not be laid off, but they won't
get pay raises either. Their job is to find a job. If they have trouble
doing so then they will train to get one, because their financial base is
eroding year by year. 

  We can now break a manager's tendency to increase his/her headcount by
requiring that the first four elements on his/her review show how well s/he
met the goals that were set for his/her organization. Any manager that does
not meet the goals is not eligible for any rating higher than a 3 except
under extreme circumstances. Personnel should resist letting a manager be
hired that has performed badly under these metrics no matter how many
people worked for him/her. 

  Digital is ideally positioned to take on these new metrics. We have
talented and dedicated employees. We still have some cash to guarantee
employment for a while. All we need to do is release ourselves from a
rewards system that shackles us to one that stimulates us. And then watch
how fast behavior will change. 

Joris Wils
T.RTitleUserPersonal
Name
DateLines
1233.1SAUTER::SAUTERJohn SauterTue Oct 16 1990 16:388
    In my experience, an individual contributor is not rated first by his
    specialty but by his breadth of experience.  If I am interviewing
    someone who has coded in FORTRAN, COBOL, BASIC, PASCAL, Oberon, Ada,
    APL and DECforms, I figure he can learn C easily enough.  If he has
    experience in data base, telecommunications, stand-alone real-time
    systems and multi-language sorting, I figure he can learn how to
    write a compiler.
        John Sauter
1233.2Maybe we SHOULD learn from other companies?MANFAC::GREENLAWYour ASSETS at workWed Oct 17 1990 10:3418
    RE: .0

    The system that you are describing is used at other companies.  It is
    called "Management by Objective".  The way it works is that the goals and
    objectives are set starting at the top and are subdivided all the way down
    to the individual worker.  It is one of Personal's responsibilities to make
    the comparisons of the goals so that people are not working at cross
    purposes.  I have seen it work at my wife's former company and it does
    appear to work.

    Lee G.

    P.S. And the answer to the question about what happens if someone is
    promoted or is transfered to a new position, is that there is an adjustment
    made to the goals to reflect the change.

    P.P.S. There was also a bonus program for those that produced over and
    above there goals.
1233.3Specialized SkillsTHRILL::WILSInterprocess Message Queueing P.O.Wed Oct 17 1990 11:128
    Re .1
    
    I consider "a list of computer languages" to be specialized skills,
    that are needed specifically for the job of computer
    programmer/engineer.
    Therefore a computer programmer is always happy to learn a new
    language, but is not as inclined to spend his/her time on teaching
    someone else how to program.
1233.4Do group members get bonusses too?THRILL::WILSInterprocess Message Queueing P.O.Wed Oct 17 1990 11:1717
     Re .2
  
>    The system that you are describing is used at other companies.  It is
>    called "Management by Objective".  The way it works is that the goals and
>    objectives are set starting at the top and are subdivided all the way down
>    to the individual worker.  It is one of Personal's responsibilities to make
>    the comparisons of the goals so that people are not working at cross
>    purposes.  I have seen it work at my wife's former company and it does
>    appear to work.
    
     Great! I had heard of that concept before, but I didn't know what it
    was exactly. At least we don't have to take a risk an reinvent the
    wheel.
    
    Question: is there room is such a system in evaluating groups? Do the
    group members get an extra bonus each if the group has exceeded the
    goals?  
1233.5KOAL::LAURENTHal Laurent, Loc: FOR, DTN: 378-6742Wed Oct 17 1990 11:548
Re .1

I totally agree with your method of evaluating a potential hire.  I like to
refer to it as hiring "the best available athlete".  However, all too often
(both in and out of Digital) I see managers restricting their search to 
applicants with a experience match.  In my opinion, this reduces the pool of
truly qualified applicants and increases the odds of hiring a mediocre person
with the "right" experience.
1233.6rtile ground for MBOGBMMKT::MCMAHONCarolyn McMahonWed Oct 17 1990 13:0416
    Please take care how, where and when you use the idea of MBO in
    Digital.  It's a logical approach that is literally laughed at in some
    circles.  
    
    Got laughed at in a meeting a while ago because I suggested
    it as a logical way to reduce risk & make decisions.  I was ashamed,
    not for me but for those high level managers in the room.  There
    attitude was that we're too unique for anything that helps elsewhere to
    help in Digital - we're smart enough just to keep on flying by the
    seat of our pants!
    
    Please ge careful.  MBO is good stuff when it falls on fertile ground. 
    A lot of the old school around here is either too smug or inflexible to
    give it a objective examination.  So please don't set yourself up as a
    target if you can see it coming.  Look for fertile ground on which to
    sow your ideas.  (And that ain't all that easy sometimes!)
1233.7A closed mind is more dangerous than any ideaMANFAC::GREENLAWYour ASSETS at workWed Oct 17 1990 14:2717
>    Please take care how, where and when you use the idea of MBO in
>    Digital.  It's a logical approach that is literally laughed at in some
>    circles.  

    I do not worry about being laughed at because it shows the intelligence of
    those laughing.

    But the bigger problem is that those who will not learn from history/others
    are destined to make the same mistakes.  Isn't that what we tell our
    children?

    To a more reassuring subject:  I do not know if there were bonuses for
    groups in the MBO I was describing but I did see that there was a definite
    relationship between the effort and the amount of respect and praise that
    the group got.

    Lee G.
1233.8Like Banks. hee heeCUSPID::MCCABEIf Murphy's Law can go wrong .. Wed Oct 17 1990 15:1710
>    Please take care how, where and when you use the idea of MBO in          
>    Digital.  It's a logical approach that is literally laughed at in some   
>    circles.                                                  
    
    It made perfect sense to me when I though someone was using MBO
    to mean Management Buy Out (as in Leveraged buy out by management).
    
    Guess I thought I was in another topic
    
                  
1233.9Reward EXCELLENCE, Not MediocrityBOSACT::EARLYSliding down the razor blade of life.Wed Oct 17 1990 18:4186
    I would submit that, to a great extent, Digital rewards mediocrity, not
    excellence. The greatest exception to this (organizationally) may be in
    sales, because sales has very objective measurements ... you  either
    made quota or you didn't. Also, the degree to which you may have
    over-achieved (e.g., 200% of quota) is also an obvious performance
    indicator. Couple this with high customer satisfaction ratings, and you
    are in good shape. The PSS part of EIS (software delivery business) is
    similar in this regard.

    The rest of the company is a big cloud in terms of metrics/goals,  and
    specifics on what constitutes success.

    As an example, I know of an employee whose performance is, and has
    always been, excellent. He came to me with nothing but "1" ratings on
    his reviews.  During the time that he worked for me, I found him to be
    "a manager's dream", and wrote these exact words on his review.

    He was goal oriented, energetic, extremely knowledgeable about his
    field, delivered dynamic presentations to customers, did all his
    reports on time, came up with great ideas, seldom bitched (and when he
    did it was for good reason), and would work his tail off to please a
    customer. (As evidenced  by the fact that he missed our Christmas party
    last year because he spent 38  hours STRAIGHT on a weekend in a bank
    where the customer was having problems.)

    This individual ended up reporting to a different organization this
    year,  through no fault/choice of our own. He clearly out-performed a
    number of other people in his line of work, yet was not a recipient of
    the "Circle of Excellence Award". Others, who I perceive probably did
    not put as much into their jobs as this person did, won the award and
    went on the trip. He stayed home. (I don't want to get into how this
    could happen, but it does, and in a case like this there is nothing you
    can do about it).

    When he confronted his management, he did so in a very sane and
    rational  way, (a lot more rational than I would have been, for sure),
    and  effectively said:

    "It's not up to me to say who deserved the award more, or who should
    win and who shouldn't. I don't want to take anything away from anybody
    who won. However, I have to tell you that I accomplished the following
    things this year (list, list, list, list), and worked an average of
    70-80 hours a week. There isn't a sales person I've worked with who
    wouldn't say that I didn't do more to help increase their CERTS than
    anyone else, and I don't honestly believe that a great number of the
    people who won this award accomplished what I did this year."

    In the discussions with his manager, he was trying to get him to 
    understand that when someone puts as much dedication into the job there
    should be some reward for going "above and beyond".

    He was willing to take on any goals for the coming year that the
    manager wanted to place on him. He was willing to work 70-80 hours a
    week  or more ... "just tell me what I have to do in order to win a
    Circle of Excellence Award next year, and I'll do it."

    Do you know what his manager said?

    1) I can't tell you what you need to do to win.

    (I have to buy this because I know how the system works in the support
    world, and you have VERY LITTLE CONTROL over who wins this award)

    2) WHY DON'T YOU SLACK OFF A LITTLE?!?!! CUT YOUR WORK WEEK DOWN TO 40
    HOURS. WHY KILL YOURSELF???!?!?!

    I might have told an employee of mine that I couldn't guarantee his/her
    participation in the Circle of Excellence, but a would NEVER suggest
    that they "slack off" or stop providing the high level of service that
    they provide to their sales reps!

    On the contrary, I would encourage them to continue their high level of
    output, and pledge that I would do whatever I could to ensure that this
    type of injustice did not occur again.

    I've seen other similar behavior from managers who had "star
    performers".  Too many have been asked to ... "chill out ...
    you're making  everyone else look bad."

    In disgust of such nonsense, 
    /se




    
1233.10it could actually be concern for the well-being of the employee involvedPSW::WINALSKIPaul S. WinalskiWed Oct 17 1990 19:1714
RE: .9

>    I've seen other similar behavior from managers who had "star
>    performers".  Too many have been asked to ... "chill out ...
>    you're making  everyone else look bad."

That's one reading of this anecdote.  Another possibility is that the manager
is trying to prevent the overachiever from burning himself out.  Eventually,
the stress of 80-hour weeks catches up with every workaholic.  There was a piece
on National Public Radio's "All Things Considered" program just last night about
how a lot of the go-getter, workaholic Japanese businessmen we're constantly
exhorted to emulate are ending up psychological basket cases.

--PSW
1233.11make a number or make an improvement - which?SAHQ::CARNELLDDTN 385-2901 David Carnell @ALFWed Oct 17 1990 23:2514
    
    Ok, here's my short note highlight from an older topic on numeric
    metrics vs. qualitative metrics:
    
    MANAGER: "But Johnny, You MUST get C's, or better yet! A's.  That's the
    way THE SYSTEM works.  That's how WE must m e a s u r e you."
    
    Johnny: "I got 'em, I got 'em.  A's, A's, A's"
    
    MANAGER: "Good boy."  Pat, pat, pat.
    
    Johnny: "Duh, fill out a job application?  But I can't read and write."
    
    
1233.12We gotta reward teamwork with bucksTHRILL::WILSInterprocess Message Queueing P.O.Thu Oct 18 1990 17:0127
    Re .11
    
    You're right, metrics don't always stimulate the right behavior. But
    you have to measure somehow. And I postulate that the current metrics
    promote primarily the WRONG behavior: i.e. individualism over teamwork
    (individual contributor) and expanding the size of the organization
    over decreasing it (manager).
    
    Re .1 & others: 
    
    I understand that you evaluate an IC on the number of skills that they
    have. I agree with that. Promoting individual achievement is good.
    However we need a way to promote GROUP achievement too! Otherwise
    people will never make a commitment to teamwork.
    
    And that way should be more than just a pat on the back, because 
    when it comes down to marketability as some of you are saying it is
    primarily "individual skills" that make the difference and to much
    lesser extent the success of their previous group. Therefore we need a
    FINANCIAL reward for teamwork.
    
    I would also like to note, that I found that levels do matter when
    you're up for hire. However someone who is among mediocre peers s/he
    will easily get a 1, but when among high performing peers will have to
    fight to get a 2. If a group of high performers is rewarded a 1 like I
    illustrated in .0 then that person will be able to get a 1 or with ease
    a 2, which is what's them due.
1233.13team reviews ? why not !BEAGLE::BREICHNERFri Oct 19 1990 08:2012
    re: .12
    "We gotta reward teamwork with bucks"
    I like it and see no reason for not doing it.
    A manger could easily write up any individual's review primarily
    based on the team's performance, with minor individual adjustments.
    
    In my long previous IC career I've seen it work the other way
    round:
    As the whole plant  blew a project, so went the "bottom line"
    for the whole plant. (my manager told me when I questioned him
    about my lousy review).
    /fred