The kaizen of communication

Kaizen is the Japanese word for "continuous improvement". It's common to come across this term when reading about Kanban.

The "kaizen of communication" is what I call the process of systematically asking for feedback. Communication is never in its ideal state because no channel works perfectly. There are always channels you could be using better, there are always team members that feel left out of a specific conversation.

Scaling communication is hard: it's common to keep using the same channels and structure for a long time and then realise that they aren't working. Over the years, I tried different techniques in the quest of improving communication. I'm now convinced that using surveys to improve communication at scale is what works best. After all, if you don't know something, asking involved parties for help can only be beneficial. A survey about the state of communication in your teams is a delicate subject, so here are a few suggestions based on my experience:

• The survey must be completely anonymous. Make sure you use a tool that supports anonymous surveys and communicate it clearly to your teams. You are not going to gather negative feedback otherwise. And what isn't working is exactly what you are after.

• Have as many closed questions as possible. Best is to ask questions that can be easily aggregated and visualised. I suggest using a Likert scale which helps create less biased questions.

• Test your questions. Coming up with clear, non-ambiguous questions is hard. Do not guess, benchmark.

• The results you share must be aggregated so that personal information can't be extracted. This is important as it's easy to fall for a blaming game. If people can infer from a survey things like "No one likes to work with this product manager", you are creating trouble not insight.

A good survey gives your team actionable feedback. Concluding from a survey that "We aren't happy with X" doesn't help. It's good to know, but it doesn't give you anything you can work with.

The way you design your questions has a big impact on how actionable the feedback is. For example, consider the following question:

How happy are you with X from 1 to 10?

Answers to this question don't help. The example is trivial, but I can't say I didn't make such a mistake. The easiest solution is to break down the question into more specific questions:

Do you find X to be the right tool for Y? What do you use X for? (Pick up to N options)

These questions help you understand whether your teams agree on the way X is used and what they actually use it for.

Sharing the results of surveys and discussing them is important. There are at least two reasons to do so:

• It helps clarify those little bits that no survey or piece of writing ever could.
• It makes the entire conversation about communication as clear as possible.

Communication is hard and never done so, at times, it's going to be a little discouraging and tiring to improve it. To make things even harder, communication doesn't improve in a day's time, so you won't see immediate results.

When I get discouraged, I stop working on the day-by-day operations of the team for half a day and do a root analysis of our most recent bug reports. I know I can often link a bug to a simple miscommunication between people. That gives me an action plan I can use to bring the motivation back.