I’ve had a few major inflection points in my career as a manager. Previously, I’ve written about managing without domain expertise. Recently, I’ve had an opportunity to reflect on another: management beyond relationships.

I have high EQ, which, for whatever reason, stands for “emotional intelligence” – I read people well and build trust quickly. In practice, this means that I’m a good recruiter and can sense challenges that individuals and teams face before they manifest. Unfortunately, having high EQ has also been a crutch that enabled me to manage much larger teams than the immaturity of my management skills should have allowed.

Since I was a member of the founding team at Dropbox NYC, almost everyone on my team was someone that I’d recruited personally. I had a relationship with each person joining my team coming in, and there was at least some implicit trust from the start. When my team was small, my team processes were unstructured, and I leveraged relationships to understand what was happening on my team and figure out how I could help. I was able to get away with managing almost 20 people this way before I hit a breaking point and had rude awakening.

In an annual Dropbox reorg, I inherited two teams based in San Francisco and expanded the scope of my teams in New York, about 30 people across 4 teams and beyond what I could manage effectively. I doubled down and spent a lot of time and effort running my old playbook – understanding what was happening on my team by leveraging conversations rather than process. I was frustrated, couldn’t figure out why, and nearly burned out.

With the next year’s reorg, I had a team of 50 people and decided to reevaluate. During one of Dropbox’s Hack Weeks, I hacked together a better way to run my organization. I created regular reviews to check in on my teams’ status, key metrics, and operational postmortems, and I designed templates for quarterly staffing assessments to track performance and career aspirations for every individual on my team. All of this process de-personalized my involvement in the team, created clearer expectations for engineering managers, and enabled managers to take more ownership and responsibility for their teams.

As my organization continued to evolve, I built some metaprocess to ensure that my operating cadence was effective. Each time that I needed to check something on my team, I asked myself whether I had a process for that (e.g. “would I get an answer to this question in our next execution review?”) and whether that process was happening soon enough to answer my question. The extra check before digging in helped me refine the agenda and cadence of these meetings and often saved me from disrupting my team.

Leveraging process over relationships enabled me to scale myself to managing much larger teams more effectively and with significantly less stress. I can still leverage my EQ, but I no longer depend on it, and everyone is better off for it!