One on One

As I started to build up the data-section of the AS-NMT I deliberately decided to avoid another level of hierarchy and started to speak with all engineers in regular 1-1 sessions. Now after 4 years of scaling the department we now reached 60 people in total and my calender basically consists of 1-1 sessions.

Would I do something different the next time - hell no!

There are already a lot of different advices what is important for a reasonable good 1-1 meeting. What I've learned over time is that there are all different. Everything depends on the person that you are speaking with. In some cases it is a simple status update - "what happened in your team in the last weeks" and in others, we do not speak about work at all (most of times about computer games :-). But all of it is totally valid and beneficial for the organization, because it is a save space to share thoughts, emotions and some times also problems that you wouldn't be aware without it.

For me the most important aspect of this kind of meetings is to build trust between the employee and the manager. If there is no trust - do everything you can by build up, or just get rid of the 1-1 and ask for a status update via mail.

Google did in their re:work initiative a lot of empirical research about the effectiveness of teams - basically they tried to answer the question why does some teams perform so well why others struggle with many tasks.
They came up with a construct which they called psychological safety which they identified as one of the key aspects of performance. In a nutshell (please read the article an the re:work pages - they are worth your time!) it means that if we human beings have the chance to learn from our and other mistakes without the fear of getting punished in any ways - we start to offer creative solutions for problems and think "out of the box". Or in other words - the performance increases. Amy Edmondson who is the creator of psychological safety offers three things to foster it - which are basically my guideline for 1-1 sessions:

1. Frame the work as a learning problem, not an execution problem. I don't get tired to rephrase in many different ways that there is no problem - as long as we do not do the same error again - which basically never happens. One more important point is there to get rid of "dreadlines" - completely pointless deadlines. Of course there is a value in setting a time frame for a given task, but introducing pressure without any given reason causes more harm than anything else.
2. Acknowledge your own fallibility. If you want your people to tell you their dirty secrets - or inabilities or failures in a wider sense - just tell them about your f**k ups that happened in your career and what you and the organization learned out of it.
3. Model curiosity and ask lots of questions. Start with - "How are you?" - but ask it honestly and deal with the things that this easy question may bring up and try to help. Ahh and one more thing - be ready to answer the question honestly yourself. One thing that one of my many managers that I had in the last decades did, which made a lasting impression on me - was to tell me that I and my needs are more important than the needs of the organization and he meant it that way. By acknowledging this simple fact he lay the ground for a deep and trusting relationship which even after switching jobs is still alive.

As all other things and advices - this is not a silver bullet and one massively important factor in this equation is time. For most cases it worked out for me - sometimes it took months, but it is one of the key aspects how I'm able to maintain an organization with 50+ people with just two formal levels of hierarchy.