People know how the medium of conversation affects the discussion itself. For example, if you have a high number of back-and-forth messages with someone over Slack, you might suggest a quick video chat to delve more deeply into the issue. More information can be conveyed, faster, with a Zoom conversation. But what if neither party had suggested a video conference? Would we realize we were trapped in surface-level conversation?

It’s not intuitive that the properties of a signal will affect its message, but they do. As different communication mediums emerge, we need to understand the implications.

Fortunately, the terms electrical engineers use do seem to translate over fairly well to human-human communication:

  • bandwidth - how much information can be conveyed in one message
  • frequency - how often can you convey it

I think bandwidth and frequency explain the relationship between email and slack.

Slack is lower bandwidth than email - while you can write a paragraph in slack, you can write much longer memos over email. But slack is intrinsically much higher frequency. It’s not unusual to slack someone 10 times a day. For two people who want to communicate normally, slack is preferable to email, because the “speed” is closer to regular conversation. For memos, where a large amount of communication is sent at once, email is preferable.

So Slack has high frequency, low bandwidth, and email has low frequency, high bandwidth.

Theoretically, you could have any conversation with any tool. Practically, which mechanism you use affects not just how much and how fast you can convey things, but what things you discuss at all.

One observation I’ve had coming out of the pandemic is that meeting up with a coworker in person is a very alien experience! We talk more deeply about the subject at hand, and I have far more “a ha” moments in person than I ever do over zoom. That seems to imply that there were fewer such moments before.

The contrast wasn’t as obvious in a world where all interactions had to happen over zoom. But now that in-person is an option, it’s clear that “live” discussions convey even more information than Zoom does, faster.

You can have more frequent zooms, but the one in-person discussion you have will achieve far more. In-person discussion, unlike any other communication form, is both high frequency and high bandwidth.

That’s bad news for remote workers, whose standard argument is that you can provide the same value from afar as you can in person. The raw math of commuting meant that working remotely seemed more efficient in terms of hours worked. But while that may be true for solitary pursuits, building software often requires a lot of interaction. Those extra hours are a lot less efficient in the grand scheme of things.

Can we measure bandwidth/frequency of communication in a conversation more directly? I think we can.

Imagine you’re in a meeting with someone over zoom about a particularly important business problem, but, in the middle of your conversation, their audio gets filled with background noise from nearby construction.

What is the likelihood that you continue discussing the problem in-depth? Do you try to explain a more abstract concept that underlies the problem?

For me, the answer is no. I’ll more commonly surf to another topic of conversation that can be covered better even with impaired communication. For example, I might talk more if I think the other person can hear me okay, but can’t speak as easily.

It’s like the old way of getting out of a phone conversation. You grab a candy wrapper and crinkle it next to the mouthpiece…“oh it’s hard to hear you so I’ll call you later”.

What about frequency? Imagine you’re meeting regularly with a team that works 10 timezones away. You’re constrained in when you can talk to the team because the working hours hardly overlap. Thus you reduce the number of meetings you have.

Framed this way, communication issues show up everywhere. How about that meeting where 10 people have their video on and 1 random guy has called in via phone? I guarantee rando isn’t speaking nearly as much in the forthcoming discussion. What about a 5 person meeting in a room, with one person over zoom?

The danger with different communication protocols isn’t about the efficiency of the communication, it’s about whether certain topics are discussed in-depth at all. If there’s some deep insight to be gleaned, it seems like it can only be gleaned in a relaxed, shared, in-person atmosphere, with time for (at least) two people to think. I don’t seem to ever get that over zoom or slack.

People who only work remotely might be just as capable, but they might be trapped in the shallow end of the pool. When everyone was in the shallow end, no one knew the difference, but as we return to in-office work, the gap will be magnified. Maybe most worryingly, coworkers won’t know to blame the communication medium - they’ll instead assume incompetence or lack of context or any of the other usual talking points. It’ll seem coincidental that the people in the office “seem to know what’s going on” and the people on the outside “seem out of touch”. And what’s worse, those assessments might be true, in the sense that the in-office person is more likely to generate unique insights than the remote worker.

Is there a good solution to this problem? I wonder if some of the newer tech, where users are in the same “virtual room” for long periods, would allow the sort of depth that is achievable so easily at the water cooler. Until then, it may be prudent to consider lower bandwidth communication as a kind of handicap - and defer deep conversations to the next high bandwidth opportunity.