One of the things that keeps us busy at bol.com is the continuous quest to figure out what else we could do to make our teams better. Even better I should write, since the vast majority of our teams are doing great already and I am very proud of them! However we also believe at bol.com that everything can always be improved. What is excellent today might not be good enough in the future. So we need to keep on developing ourselves and that is also applicable for our teams.
A new idea
Over the years we have been very successful in becoming better by changing our ways of working; by organizing ourselves differently; by recruiting with a strong focus on our culture; by investing in new technologies or tooling; etc. We are proud of everything we have achieved and have celebrated our successes! And this also gives us a lot of confidence that we can keep on improving our teams in the future. In this post I like to share a new idea some of us have and which could be another big thing.
At this moment this is an idea, a hypothesis, a concept and not a best practice (yet..). We are looking for valuable input to sharpen and mature this concept and are doing that by reading, talking, discussing and sharing in order to collect feedback. And we are experimenting with this on a small scale.
A perfect team
All our teams are committed to create more value for our customers, our partners or our employees. Being able to do this easier, faster and better is what we see as an improvement. It is crucial that these are sustainable improvements. So that we are able to benefit from these improvements not once, but continuously in the future.
This quest for continuous improvements is a difficult journey and we will never be able to reach our ultimate dream: the perfect team. But the quest inspires and provides us with challenging problems which we want and need to solve. We know that high performing teams are better than good teams. Their impact is at least double the normal impact!
A dream team
We will never be able to find or form the perfect team, but dreaming about this is something we can do. So we dreamed, pondered and discussed. This process triggered us to think about dream teams. The real dream teams that have existed in the past!. At bol.com (of course… ) but in this case we are referring to sports teams. Remember the USA Basketball team at the Olympics in 1992. Or the Football team from FC Barcelona in 1992 (great year in Barcelona). And again in 2009 (the team lead by Pep Guardiola). Also Ajax in the early seventies can be seen as a real dream team, winning the European Cup 3 times in a row. We remember these teams and – often with melancholy – we think (dream…) about the great moments which they have given to the world.
At bol.com this triggered the discussion on what we could learn from sports teams and we came up with an interesting hypothesis: to improve teams we don’t need to improve teams…
To improve teams
We are not trying to compare our scrum teams with these sports teams. Instead we are trying to learn from a professional sports organization. We are fascinated by the fact that over the years sports teams have invested heavily in all kind of supporting roles around the team. And they continue doing so, as it has proven to help them in their ambition to become more successful.
When Ajax won the European Cup for the first time in 1971, they had a great team and one unique coach. When FC Barcelona won the Champions League in 2015 over 15 persons were part of the technical staff. This support staff consisted of a head coach, several assistant coaches, individual trainers, analysts and even a psycho-therapist and a psychologist. And they have probably learned from American professional sports clubs, who realized decades before that it pays off to invest in top coaches, mental trainers, fitness experts etc.
These top sports clubs invest in talents around the sport team to enable the talented players to maximize their performance. They believe, and obviously gathered proof, that investing in expertise around the team will increase team performance. The players within the team can then focus on that what they are really good at: playing the game! We could only draw one conclusion: to improve the teams we don’t have to improve teams but instead have to improve the support organization around the teams!
Why might this work?
At bol.com and probably also at other organizations we expect more and more from our teams: we don’t only expect a scrum team to build great software, but we also want them to challenge the stakeholders when proposing new innovation. We want the teams to deploy software continuously to production, we expect them to monitor the behavior of their services running in production. We want them to pick up the phone (and act) in case of an alert. And we expect them to stay up-to-date on what is happening in the tech community. This can be overwhelming for the team members and is not always easy. The obvious question which came up is what we could do to make this a bit easier. And the (as) obvious conclusion is to help them to be successful in all these different tasks.
At this moment we have several scrum coaches at bol.com to help the teams in the scrum process, we have software architects supporting them in the architectural challenges they encounter, we have a space reliability team to support the operational activities etc. But we are wondering if this is sufficient and sometimes think we need to add more support roles outside the team for the teams to be able to excel.
Recently we have started an experiment with a role which we have named IT-catalysts. The catalyst is an expert who can oversee a broad domain from both a technical and a functional perspective and who is helping to implement complex multi-service epics. This person is working closely with a small group of teams including the product owners and other stakeholders. He or she is supporting them on a daily basis.
Also we have a test architect now, a relatively new role at bol.com, who is helping teams to take the right approach for their test strategy. And we have started to experiment with information analysts working closely with 2 or 3 teams and supporting these teams on a daily basis.
Why is this difficult?
These supporting roles could be seen as overhead and hence not adding direct value for our customers, partners or employees. This seems to be contradictory to how we have approached things in the past, as we try to be lean and focus on growing where there is direct impact. It seems much more logical to add a new team member to a team, expecting the team to go faster and deliver more value. Or to create a new team to build new innovations. But more is not always better, because more team members could also lead to more discussions. Or to more dependencies, or alignment meetings etc. And maybe the team balance is already very good and changing this (with all good intentions) might do more harm than it would give us. But if we cannot or don’t want to change the team then we can only look into options outside the team to improve them.
At bol.com we have done some experiments which have proven that adding someone outside the team might sometimes be the best solution to improve team performance. We have piloted with this in several areas and we have seen successes. As most know, it is not so easy to measure team performance and we are trying to solve this by collecting evidence that a team has improved: happier team members; compliments from stakeholders; less production issues; more stories completed by the team; etc. Sometimes the evidence is more factual, sometimes more exemplary. The overall picture has to be positive and at bol.com this has been the case. Our preliminary conclusion is that to improve the teams we don’t need to improve the teams. And we should continue to do experiments to really prove that!