big

NOOP.NL | The Creative Networker

The Merits System

06/02/2013

It is clear that the annual bonus system doesn't work. A tremendous amount of research says that it demotivates people, destroys collaboration, and causes dysfunctional behavior among managers and employees.

What also doesn't work is the flat system, where everyone simply gets the same compensation or bonus. This also demotivates most people. And it makes organizations inflexible in times when agility is needed.

What we need is a merits system

Merits system color

5 Criteria for Fair Compensation

Coins colorIn an agile organization, working in an uncertain environment, I believe workers should have a steady salary that is predictable and slightly conservative. On the other hand, they should also get extras depending on the unpredictable part of the environment. Both salaries and extras should be brutally fair and based on merits, not equality. This has led me to suggest the following practical constraints, for better compensation systems, based on the five problems listed in an earlier post:

  1. Salaries should be expected, but bonuses should not.
    Always keep it a surprise. When bonuses become frequent and anticipated, they ought to be converted to regular salaries.
  2. Earnings should be based on collaboration, not competition.
    When determining how much people should earn, the main criteria should be their contributions to a common goal, or shared purpose.
  3. Peer feedback is the main performance measurement.
    Contributions to a shared purpose are best detected and evaluated by peers, not by managers. Only the whole system knows all details.
  4. Use creative thinking to grow the compensation system.
    Expect that people can (and will) game any system, and tap into that creativity by inviting and supporting it, instead of driving it out.
  5. Use compensation to nurture their intrinsic motivation.
    Make money a reflection, not a competitor, of people’s curiosity, honor, acceptance, mastery, and all other intrinsic motivators.

Of course, implementing these suggestions for a compensation system is not a trivial thing. But I discovered different ideas that seem to work pretty well for various creative organizations, and which turned out to be quite compatible with each other, and with the science of behavioral economics [Ariely, Predictably Irrational].

A Better System

The elevator-pitch-version goes like this…

  1. You start by setting up a virtual currency, such as credits, points, or hugs. And every month every employee gets an equal share of the virtual currency.
  2. Employees are then required to give all their credits to co-workers. They can use any criteria they want, as long as they don't keep any credits themselves.
  3. Employees accumulate earned credits (received from co-workers). At random moments they can exchange credits for real money, using an exchange rate.

This is a payment system based on merits. It satisfies all criteria for fair compensation:

  • Big bonuses cannot be anticipated.
  • Peer feedback is the main performance measurement.
  • Earnings are based on collaboration, not individual performance.
  • Creative thinking is used to game and tweak the system.
  • The earnings reflect purpose, mastery, status, etc…

Run a Safe-to-Fail Experiment

Hugs colorMoney and emotions are tricky things and therefore any system that involves both will have to be set up in a way that is safe to fail. With small increments (such as weekly or monthly experiments instead of quarterly or annual outcomes) the feedback cycle is shorter and people will learn faster how to improve the system. The use of valueless virtual currency instead of real money will allow people to experiment more comfortably, and it will be easier for them to decide that a chosen path is not working, and change direction, or start from scratch. We must also realize that creative people will game the system. But this creativity can be exploited to grow a more resilient system. The short iterations and valueless currency should help people to adapt to each other’s strategies, and management can tweak the constraints, all in favor of collaboration and working towards a common purpose.

If you are interested, I have a 20-page article about Merit Money available for download.

Merit-money-mini-150This text is part of Merit Money, a Management 3.0 Workout article. Read more here.

Subscribe to my mailing list and be the first to receive future articles!

This article is written by on in Workout. Jurgen Appelo is at Happy Melly. Connect with Jurgen Appelo on .

This article was posted in:

  • http://profile.typepad.com/mikemallete Mike_mallete

    Great writeup as always! Though I do have a few opinions on the matter.
    From the same book by Ariely, he mentioned about the phenomenon of more cheating within only a single step away from money. And I do see “merits” as you describe as such. I could be contextually wrong of course. It may fulfill most of the criteria you set. But tying points to money/reward whichever the intent is will obscure the main reason for giving such: honoring the other person, and giving feedback that he/she is doing a good job. And instead have money/reward as the main focus. I would rather disconnect monetary reward from the merit system. And instead make it more about allowing other people to publicly recognize someone. Like stickers, fun toys they can display on their desks, etc.
    Now for the real goal of this system which is distribution of rewards. I don’t see a problem on even distribution pending some rules: you have a culture of feedback within the whole organization, rewards are distributed per team, and allow the team to figure out how they will distribute them among themselves. I would lean more towards allowing the teams to figure it out rather than create a system for them :-)
    Most of this I am saying from my limited experience.

  • http://profile.typepad.com/jurgenappelo Jurgen Appelo

    Thanks Mike, these are good comments. But I don’t think this really addresses the issue.
    First of all, how do you decide how much money goes to which team, assuming they have different sizes and contributions? Someone needs some criteria here.
    Second, how do you deal with individuals who don’t work on a team? Again, some criteria are needed.
    Third, if you leave the decision (for how to distribute money) to a team, they have different choices. They can use a bonus system (which is bad). They can use a flat system (also bad). Or they can use some kind of merits system for the distribution criteria. If they do this, I suggest they read my article. :-)

  • http://profile.typepad.com/mikemallete Mike_mallete

    Thanx for the reply Jurgen!
    Great points. Let’s see if I can address them.
    How much money goes to which team is indeed situational. But I would step back a bit further. How does the company know how much to give as reward as a whole? If it’s due to end of year excess in target for total income, shouldn’t be a problem evenly distributing per team. If it’s per project basis, perhaps that’s on a per team basis. For situations like these though, I would prefer the former. As it will be unfair for teams who don’t have a choice on which project they will work on, the type of possible profits out of it, etc. Very simplistic criteria perhaps. But I believe will enhance the social contract and peer pressure on a team level.
    With regard to dealing with individuals who don’t work for a team, I don’t think a merit type system would address that either. You can formulate an alternative, with the team at the center of it.
    For the third point, I don’t think that leaving the decision to the team is necessarily bad. In the end, it’s their choice. And usually, given autonomy, they will stick with it or fix it given the chance. I believe the wonders of self-management goes beyond software engineering.
    An example of this is when we tried to distribute rewards on a percentage basis. The interesting phenomenon is that those with the higher salaries actually wanted it to be flat rather than them get the higher reward. That’s the last time I would want to create a system for them :-) They are way smarter than me.

  • http://profile.typepad.com/jurgenappelo Jurgen Appelo

    “shouldn’t be a problem evenly distributing per team”
    Of course that’s a problem. It is unfair if the teams are different sizes. It’s unfair if some teams only existed half a year. It’s unfair in many cases. It is you who is creating a rule here, and it won’t work because people are smarter.
    “With regard to dealing with individuals who don’t work for a team, I don’t think a merit type system would address that either.”
    It certainly does. Individuals outside of teams can be rewarded by team members just like normal team members. Because of some great service the individuals offered to teams. Or just because they’re great friends.
    “I don’t think that leaving the decision to the team is necessarily bad. In the end, it’s their choice.”
    I agree. That’s why I suggest to give every person exactly the same amount. And in the teams they can decide what to do with that.
    “That’s the last time I would want to create a system for them”
    Well, if you correlate money to team boundaries, that’s exactly what you’re doing: you’re creating yet another system. A system of team boundaries.
    I think you should leave it to the individuals to decide what their boundaries/personal networks are.

  • http://profile.typepad.com/fischergordon Gordon Fischer

    Love this method. 7 years ago we tried something like this at a startup but it crumpled for reasons you’ve corrected above.
    One question – Is there software you know of to handle distributed teams? It’s hard to give someone a hug over 300 miles. :)

  • http://profile.typepad.com/jurgenappelo Jurgen Appelo

    Sorry, I don’t know any tools.
    Hope to get some suggestions too…