The Agile Forest

I posted recently about my fascination with online role playing games and how some of what intrigues me is around the area of incentives.

This has had a bit of attention in the Agile space with Daniel Pink covering off motivation in his book ‘Drive’ but for the quick versions you can see his TED talk or checkout an animated version.

The general gist of motivation as explained by Daniel is that if you reward people based upon speed of task completion, the stronger the reward the more likely to have an inverse productivity effect. The reason? Work where any element of innovation is required, ie the solution path is unknown, is stifled by rewards.

Now Daniel says that this is due to the fact that a reward or incentive is promised. I wonder what the impact would be if there was just a push from a time perspective – ie rather than encourage a reward, try and get it done faster by saying the last team got the candle test done in ‘x’ seconds and put a big huge clock in the room counting it down.  I think that the same problem would occur, but would this be because the time pressure is stifling innovation or because the challenge itself is creating a human instinctual reward – “If I do it faster than that last group then it shows that I am smarter and if I prove I am smarter than I feel better.”

Daniel says that you are better to remove reward systems and replace them with innovation systems eg FedEx days to create motivated employees.

So how does this fit into online games? What motivates millions of gamers to play? Well naturally it is fun. FedEx days appeal to many because they are fun. Is it purely that simple? Make work fun and you have fully motivated and engaged employees? I’d say fun goes a long way to motivated employees but it isn’t the whole story.

There are many different elements of gaming that is not fun. ‘Levelling up’ being the one that comes foremost to mind. When you start playing the game, as with real life, you are an inexperienced virtual person. Through learning and using your skills you become better. Completing missions or ‘quests’ would earn you experience. Defeating creatures would also earn you experience. Each time you levelled up the experience required to complete the next level would increase.

Back in the ‘ol days of online gaming levelling up pretty much was a majority of the game, taking half a year to get one virtual player to the maximum level possible. There was a lot of push back from players and consequently this has been slimmed down over time to a few weeks to under a month assuming a player focussing 25 hours per week. (Sounds familiar – waterfall to iterative anyone?)

Individual’s rarely cheered or got a thrill when they completed a quest or defeated a creature. But levelling up was a big reward. Other players in the area would congratulate you, your virtual family (guild) would congratulate you, you would also receive a new skill or an improvement on an existing skill. This is likened to getting a very small promotion frequently. In Agile, quests would be like ‘done’ stories, where as levelling up would be more like a release or maybe an iteration showcase.

The second ‘grinding’ element of online games occurs once you reach the maximum level and begin having to save money to buy clothes or other virtual related items, for example, vehicles or mounts. This element reflects a capitalist life all too much where you must complete the same monotonous quest every single night in order to receive money. This sort of re-occurring capitalist related behaviour is already starting to be phased out of online gaming and replaced with ‘weekly quests’ or slightly random regular quests in order to be more engaging.

The last of the less motivating elements of online games is around a concept of ‘raiding’ and ‘downing bosses’. ‘Raiding’ involves getting together with a bunch of other people, usually from your virtual family, and working as a single unit to defeat a complex non player character (boss) that is a few levels above you. This often involves working in teams of 5,10, 20, 25 or 40. Like software development, the larger the team, the greater the complexity required to herd cats. Conceptually you can think of ‘downing bosses’ as a large point Story and completing all bosses in the dungeon as an epic or release. Bosses can take anywhere from a few hours up to sixty hours for the whole team to complete. Whole dungeons of bosses can take a few to several months of work to complete.

I remember fondly one particular boss that required utmost perfection. Not just of defeating him, but defeating those that came before him. The less you failed, the higher the rewards. It was an interesting concept and probably is the key tie to quality. But for this boss we spent around forty hours before he was finally defeated. The way that he was coded meant that each person had a role they had to fulfill and in order to defeat him it required everyone to fulfill their role at the right point in time with 100% perfection. Because getting together as a huge team took quite a lot of logistical effort we probably only spent eight hours a week on this fight and consequently it took five weeks to progress. A five week epic. And what happened when he was defeated? Riotous cheering amongst every team member. It went on for minutes. The high lasted for days. When was the last time you cheered at the end of completing an epic? When was the last time you were on a high for days for doing an iteration?

The complexities are the same. Working together as a whole team to get a particular outcome. So why is the engagement so different between the two? What is the reward in online games for such an effort? It often isn’t about the gear or financial rewards you get, it is about the bragging rights, it is about the sense of achievement. We can’t compare apples with apples here when we deliver a Story. No one else in the world is delivering the same Story to be able to say ‘Hey we delivered As a Call Centre Operator I want to see the all the products that my customer has in a single view, have you done yours yet?’ or even ‘We did it in 8 days, how long did it take you?’. We need to make sure Story completion is regarded as an achievement.

So how can we make work more fun and more rewarding looking at online games:

1. On demonstrated skill improvements make a big deal. Go out of your way to congratulate those that made it.
2. Go out of your way to celebrate a release to production and epics being completed. I’ve seen release parties for Waterfall. I don’t think I have ever seen this for Agile, we need to do it more.
3. Regular, repeatable activities that require no level of innovation are not fun. Effort needs to be expended to switch it up more or vary the activity somewhat to keep an employee engaged and motivated.
4. In a Kanban like system where we aren’t celebrating at regular timed intervals. This means we need to concentrate harder at celebrating success when an epic is completed.
5. Challenge has to exist but not so much that a person feels overly stressed. Getting this balance is important, challenge encourages the need for innovation; innovation makes the mental part of ourselves feel rewarded.

Lastly let me leave you with this one thought. What if we weren’t paid an annual or fixed salary. What if we were paid on the Stories that we delivered as a team as a percentage of the Story’s actual returned value? What behaviour would it drive? There would be potentially lots of really bad behaviour out of this, but the one thing I can’t take my mind off thinking is that finally we would focus on real value. The metrics we would have to predict value would be a lot more thorough. Daniel’s findings suggest this would be a dreadful thing as tying a financial element to the work will have a negative effect, but if you have a fixed definition of quality and no time restrictions would it really not empower innovation? To me it would still drive innovation because we would be directly tied to ensuring that our delivery will result in optimal outcome of value for the organisation.

Read Full Post »