Building a culture of business owners and incremental innovations

Don’t ask for permission — ask for advice

In a high growth environment, osmosis and direct mentoring is not always enough to shape culture. To be deliberate about culture, I find it useful to write down a strawman, as an anchor to provoke debates with my team. Here is a short piece.

Context: in my teams, to enable individuals to be autonomous while fostering collaboration, I delegate quarterly objectives to squads of 2–4 people.

Transferring ownership without losing the momentum in the delegation process

Act as a business owner, which all of us are, and more specifically as owning the team quarterly objectives.

Teams owning objectives, or individuals owning tasks, are often handed not only with a goal, but also a vision of how that goal could be reached, along with perhaps prototypes that were made to derisk or scope the task.

Those visions or prototypes are valuable, and should be discussed in depth with their originators so they are well understood. But they are not gospel, and it is essential we challenge them in a constructive way, envision the future, and derisk what’s not clear in either the implementation or the requirements, at any step of the process.

For the best outcome, this needs to be balanced with focus and coordination. If we only explore without delivering, we’re exploring around a single point, like a Brownian motion without drift. So we need cycles of exploration and execution to maximise our learning.

When we decide which option to explore next, even if your option hasn’t been selected, and even if the decision isn’t the best one, we all need to stand behind the decision and push it to the next iteration cycle so we gather new information.

In optimisation theory, this is similar to the gradient descent algorithm, trying to point towards a global optimum while having only local knowledge at hand: the best algorithms aren’t the ones that maximise the search time from local knowledge (e.g. computing lots of derivatives and heuristics from lots of records), but those that balance this with more iterations — stochastic gradient descent being an example on an opposite extreme.

In a nutshell: keeping your eyes on the end goal, challenging each other, balancing exploration with execution and iterating often.

How does that translate into rules or processes?

It doesn’t make sense to hire smart people and tell them what to do; we hire smart people so they can tell us what to do.

Instead of rules and processes, many of which become stale pretty much immediately, we work towards shared goals and with a shared culture, allowing the bright minds in Vortexa to create value, coherently.

Don’t ask for permission: you are the owner — ask for advice.