Are presence dots helping async work, or just rewarding visibility?

I manage a distributed engineering team, and I’m increasingly unsure that Slack-style presence or “last active” indicators improve asynchronous collaboration. They show activity, not concentration, progress, or willingness to be interrupted. Across time zones, a green dot can quietly become an expectation to respond, which blurs coordination into surveillance and rewards performative availability.

I don’t think presence is useless. During an incident, knowing who might be reachable immediately can help. But for normal work, I’d rather set team-level response agreements—what “normal,” “urgent,” and “incident” mean—and require handoff notes with current status, risks, next action, owner, and response window.

For example, after an overnight deployment, the next engineer should be able to continue from that note without inferring responsibility from activity dots. Have your distributed teams replaced presence tracking with clearer asynchronous commitments? I’d genuinely like counterexamples or disagreement.

The incident exception is the important one. Presence is a weak signal, but weak signals are still useful when the alternative is paging everyone. I’d keep it available for incident channels and explicitly forbid using it as a routine performance proxy.

I agree with the direction, but I’d be careful about claiming the indicators themselves cause performative availability. The stronger claim is that they can be interpreted that way, especially when response norms are unstated. That is enough reason to define the norms.

The interface teaches the social rule. A green dot looks objective, so people treat it as permission to interrupt even when it only means the app detected activity. Handoff templates make the meaningful state visible instead: what is blocked, what is next, and who owns it.

Counterexample: on a small team with overlapping hours, the dot can prevent needless calendar meetings. I’ve seen people use it as a lightweight “is this a bad time?” signal, not a demand for instant response. The problem may be managerial interpretation rather than presence itself.

There’s also a privacy angle. A timestamp creates a trail whose usefulness is easy to overestimate and whose context is easy to lose. If the operational need is incident reachability, publish an on-call or escalation state instead of collecting broader activity clues.

Response agreements only work if the categories are genuinely understandable. “Urgent” becomes everything unless examples and escalation paths are written down. I’d pilot the handoff format first, then measure missed dependencies and unnecessary interruptions rather than online time.

A green dot is a door sign, not a service-level agreement. Put the agreement somewhere people can actually follow, and the dot becomes optional decoration. I’m not sure most teams need another dashboard for knowing that.