Vivek Ganesan bio photo

Vivek Ganesan

A techie, an author, a coach, a learner, a blah, blah and a blah!

Email Twitter Facebook Google+ LinkedIn Instagram Github Stackoverflow Soundcloud

Ads by Google


Blameless Continuous Integration


While I was reading the book Managing for Happiness written by Jurgen Appelo, I came across a concept called Merit Money. Jurgen Appelo says that we can improve ‘good’ behaviors and collaboration among people in an organization by introducing a Merit Money system where anyone in the organization can award virtual currencies, whose value fluctuate over time, to anyone else who exhibited a good behavior. Everyone will be allotted an initial seed amount of some virtual currencies to start with and each quarter or year, they can be redeemed for a physical reward or money.

The ‘Merit Money’ concept shows striking similarities to the ‘Open Coin Bank’ initiative that my team tried two years ago, while I was serving as a Scrum Master for a team of 9 people. This initiative was to reward the ‘collaboration’ among team members and not the competition. We tried it at ‘team level’ and not at organization level.

What did we do in ‘Open Coin Bank’?

I do not remember where my team got that inspiration from. We created a excel workbook named ‘Open Coin Bank’ and anyone in the team can access and edit that workbook. This spreadsheet contained three tabs which contained information as follows.

Tab 1 - Rules of Open Coin Bank

Screen Shot 2017-04-19 at 3.55.32 PM.png

This tab introduced the idea to anyone new and set some rules for the bank. Here are the rules.

  • This open coin bank is ‘open’.
  • All members will have unlimited access to edit it.
  • The “Summary” tab contains the current balances of each member
  • The “Transactions” tab contains all transactions of coins between members


  • Everyone gets 1000 coins as starting investment from the Eye of God during joining.
  • People can award coins to their buddies on receiving any help or for appreciation based on their own discretion. No Force.
  • The person awarding the coins will update the excel sheet with a new transaction and will update the both sender and receiver’s balances in appropriate tabs
  • The Eye of God will reward 200 coins to a member on 5 nos. of transfers flowing to his account
  • On reaching 2500 coins, the transaction history of the member will be assessed and 2500 coins will be converted to rewards based on assessment of usefulness of contribution
  • On detecting malpractice, the member detecting the malpractice can claim 1000 coins from the person who is caught doing malpractice.
  • PO/Architects/Leadership can recommend some coins from Eye of God for any Bonus deliverables worked on by a person.
  • Bonus deliverables should not take more than 1 hour for coding + positive and negative testing
  • Bonus deliverables are anything that is not expected but delivered. They can be small bug fixes, usability fixes.
  • Each release cannot have more than 4 bonus deliverables


  • Anyone who is caught doing malpractice have to pay 1000 coins as fine to The Eye of God apart from giving 1000 coins to the person who caught the malpractice.

Tab 2 - Summary of each participant’s current balance

Screen Shot 2017-04-19 at 3.45.34 PM.png

Tab 3 - Transactions

Screen Shot 2017-04-19 at 3.50.10 PM.png

Positive Outcomes

  1. People started giving coins to other teams’ members (who were not in the program initially) – Developers gave coins to support team members and some support team members gave their earned coins back to developers who helped them sometime.
  2. Since peers were handing out coins, some behaviors which do not get often rewarded from higher-ups, like clearing tech debts, writing Junits, etc. started getting rewarded. People started getting conscious about craftsmanship.
  3. Customer support team got help easily from developers in case of any need since that was an opportunity to get a reward.

How long did this last?

Given that the system was started roughly two years ago in June 2015, can you guess how many transactions would my team have done so far? Some thousands? The last time I checked, the number of transactions so far was 17. Yes. Just seventeen. Last transaction was on 4th of August, 2015. Thus, the initiative lasted for a full one-and-a-half months before dying its natural death.

But, isn’t this similar to the Merit Money system of Jurgen? Yes but this system has some not-so-good differences too. There were multiple dysfunctions in our system that I can find in retrospect, AFTER reading Jurgen’s version of the same. I realized that this open coin bank looked similar to Merit money only on the surface.

  1. Redemption was done only when people accrued 2500 coins so the time of redemption is not fixed. There was no common point in time when people could look forward to celebrate together.
  2. Since people were giving roughly 100 coins per transaction, it would take too long to get 2500 coins in anyone’s account, taking into account the fact that one has to give away coins too. If the redemption was done at quarter end similar to Jurgen’s example, it could have taken care of people’s motivation because everyone would get something at that time, irrespective of how much they had earned.
  3. We acknowledged that there could be malpractice and heavily penalised it. There is a chance that people could have felt untrustworthy.

If I try to evaluate this Open Coin Bank initiative against Jurgen’s checklist for rewards, the deficiencies in this otherwise good looking system become easily visible.

Rule Did ‘Open Coin Bank’ follow this rule?
Don’t promise rewards in advance No. The system was an open promise for a reward for anyone who got 2500 coins and none for others. Moreover, there was no ‘chance’ element for the reward.,In short, the rewards were not dependent on the environment.There was no ‘roll of dice’ that could decide one’s encashment without affecting the accruals.
Keep anticipated rewards small Can’t say. Since there was no guidelines for what 2500 coins were worth, we cannot say whether people had ‘big’ rewards in their anticipations or not. But, one thing is for sure – the motivation for using this open coin bank did not stay for long.
Reward continuously not once Theoretically yes. Practically no. Since the initial goal was too high (2500 coins), it was difficult to envisage the second instance of encashing the reward, leave alone the first instance.
Reward publicly not privately Yes. However, we could have made this rewards more public by encouraging physical ways of giving the currency rather than quietly entering in an excel sheet.
Reward behaviors and not outcomes Partly Yes. There was a clause of ‘bonus deliverables’ for which PO/Architect/manager can recommend coins.
Reward peers, not subordinates Partly Yes. There was a way in which one can earn coins from their managers too.,One big loophole was that there was no ‘upper limit’ on how many coins the higher-ups can give from the Eye of God to anyone. This promotes people trying to please managers as it helps their performance appraisal in addition to the coin encashment.

Were these differences the ones that contributed to the death of Open coin bank? I do not know for sure. But, I now have some good improvements to the system that can be tried for the next cycle. Thanks to Jurgen for the pointers and to my team which helped my thought process by trying this experiment!