Cross-functional teams at carwow
Since the beginning of carwow I dreamed of a tight group of experts from different disciplines working together for a common goal. Since our first few hires I’ve always been aware of how things would change with 30 employees, 100 employees, 500 employees, etc.
One of the reason I co-founded carwow was to build a better type of company. In particular I wanted to achieve the following objectives:
- Clear decision strategy based on data (eg. No politics)
- Low risk aversion in order to facilitate innovation (e.g. move fast and break things)
- Flat organisational structure
I believe that small autonomous interdisciplinary teams are the way forward to maintain the above principle as we scale.
So now that we are almost 100 people working at carwow we’ve started feeling the pain.
We lacked focus, clear internal communication and to an extent we struggled with prioritisation (and decision making).
The first change we’ve introduced is Objectives and Key Results (OKRs). I might speak more about them in a future post; the web is full of information about them. I’ve found the book “Radical Focus by Christina Wodtke” invaluable.
OKRs helped us first define the overall company strategy and then helped each team decide the top 2–3 objectives to focus on in order to achieve the company objectives. But it also highlighted a problem… we couldn’t find a way to separate objectives of the data team, dev team, design team, etc.
Finally something clicked, now was the time to create real cross-functional teams.
I was originally reluctant to introduce them in the past; mainly because most of the examples I had seen in larger organisations showed teams divided across different products. This may make sense when you have 100s of engineers, like spotify, to have a team focussed on “recommendations”, or a smaller part of the system.
It didn’t make sense for carwow. Instead what worked for us was to align the teams across broad objectives.
Our company value chain is summarised by the following steps:
1) Attract potential buyers effectively
2) Get these potential buyers to get in touch with the right dealer for them
3) Make sure that dealers offer a good service and a competitive offering
So we now have three product teams that have a mix of product managers, designers, data analysts/scientists, developers and some occasional specialists like copywriters, SEO or email experts. They have a clear objectives and they are strongly aligned with other parts of the business (e.g. User acquisition product team works closely with marketing and content, Supply product teams works with dealer account managers, etc.)
It wasn’t an easy change to implement, and there was some resistance. We’ve spent the two weeks before “the moving day” and the two weeks after making sure everyone in the organisation was OK with it.
The change was quite quick for many, but that’s the way we do things at carwow; we debated, decided and implemented within 2 weeks!
There is still lot to figure out, but we’ve felt was best to feel the pain as soon as possible and iterate from there (it sounds like our code deployment strategy).
After the initial pains, we are already seeing some benefits. For example a faster feedback loop between product managers, developers and data. No need to prioritise developers to support the data team. If a data analysis task can be simplified by some help from developers, we are now just doing it - Because we are in the same team, and we have the same priorities.
The new structure is also helping us with highlighting some existing problems, for example work that was deemed important by some team but when looked at in the context of the overall company goals, wasn’t.
And most important of all we are going through the phases of team formation: Form — Storm — Norm — Perform. We are still at ‘Storm’ but we can see the light. So far the main conflicts have arisen from the different way the previous teams approached planning. The new product owners are used to fast iterative work, some teams were used to longer iterations.
The next step for us is to facilitate the bonding of the new teams, resolve the conflicts, and get to the ‘Norm’ stage as soon as possible.
We have clear success metrics, and I’m looking forward to demonstrate to the company the worth of each team. With success we should see more autonomy, with more autonomy I hope to see more innovation.
Interested in making an Impact? Join the carwow-team!
Feeling social? Connect with us on Twitter and LinkedIn :-)
