Yesterday’s post looked at the fast responder — the person work routes through because they answer well and answer quickly. This one looks at the same situation from the other side: what it is like to be one of the people asking.

From where you sit, you sent one message. It was short, it was polite, it was clearly scoped. Nothing about it felt like an imposition. What you cannot see is that it arrived alongside eleven others, that four of them are also short and polite and clearly scoped, and that the person on the other end is currently deciding which of the twelve to answer before their next meeting.

Invisibility Is the Mechanism

A queue you can see behaves differently from a queue you can’t. Stand in a visible line and you make real decisions: whether to wait, whether to come back later, whether this errand is worth forty minutes. Nobody in that line thinks they are the only person there.

Direct messages hide the line completely. Every requester sees a conversation of one. The natural inference — I am a small, reasonable ask — is correct in isolation and wrong in aggregate, and there is no feedback anywhere in the system to correct it. You never learn that your question was eighth. You just notice the reply came slower than usual and quietly wonder if something’s wrong.

This is why “just tell people you’re busy” fails as a fix. It converts a hidden queue into a hidden queue with an apology attached. The requester still cannot see the shape of what they joined, so they still cannot make a different decision — and the only lever they’re left with is to ask again, more urgently.

What Escalation Actually Signals

When people ping twice, mark things urgent, or bring a question to a channel after a direct message went unanswered, the usual read is impatience. Sometimes it is. More often it is a person operating with no information about their position in a line, using the only instrument they have to probe it.

Escalation rate is a decent proxy for queue opacity. If a team escalates constantly, the first thing to check is not whether people are being pushy — it is whether anyone can tell what is already in flight and roughly when their turn arrives.

Making the Line Visible

The fix is boring and it works: put the requests somewhere everyone can see them.

A shared channel where questions are asked in the open does most of the work by itself. Not because more people answer — though some will — but because the ninth asker can scroll up and see eight others. That is enough to change behavior. Some will wait. Some will find their answer in the thread above. Some will realize their question is the same as someone else’s and add a reaction rather than a new message.

Beyond that: give the queue a rough shape. “I’m working through requests from Tuesday” tells someone more than “sorry, swamped.” A visible list — even a crude one, even a pinned message — turns an unknowable wait into an estimable one, and estimable waits do not generate escalation.

And when the queue is visible, something else becomes possible that never was before: you can look at it and see what it is made of. Twelve requests that are all variations of one question are not a staffing problem. They are a missing document, or a bad default, or an interface that surprises people at the same spot every time.

You cannot see that pattern in twelve separate direct messages. That is the real cost of the invisible queue — not the waiting, but that nobody ever gets to look at the whole thing at once and notice what it is telling them.