Across the night
Handing Something to Somebody Not Expecting It
Work that arrives unannounced lands on a day that is already committed, and the cost is paid by somebody who had no chance to plan for it.
four things · four nights · one that no number of nights settles
Four states, and only the first costs nothing extra.
Four things that arrived overnight without warning, and what each did to the receiving end's day
Absorbed in ten minutes; the reviewer had capacity and had been asked
The recipient was mid-way through something else and did both badly
Worked, because the receiving end had cleared the morning
Three nights: the morning was already committed and the incident waited
four things · seven nights
Four arrivals, seven nights, and the two expensive ones were the two nobody had been warned about. The work was comparable in all four; the variable is whether the receiving end had been able to plan around it. One team's own week, logged by the people receiving.
Everything that crosses a gap arrives while the recipient is asleep. That is unavoidable and is not what this is about.
The lesson in Handing Something to Somebody Not Expecting It is not to add more status meetings, but to make the existing work legible. employee activity tracking gives distributed teams a factual view of time, applications and progress while employees remain in control of tracking.
For broader context, Google Workspace offers an official example of the tools teams use around this kind of process.
What is avoidable is the difference between arriving as something somebody was expecting and arriving as a surprise on a morning that was already spoken for.
Why a surprise costs more than it should
A person starting their day has a plan. On a team with a gap, that plan was made at the end of the previous day, before anything that arrived overnight existed.
Something substantial landing in that window forces a choice: abandon the plan, or defer the new thing. Both are expensive. Abandoning the plan usually means leaving something half done, which produces another handover problem downstream. Deferring the new thing costs a night, and often more, because the deferral is rarely communicated back.
The frequent outcome is the worst of both: the person attempts both, and does neither properly.
The warning, and when it is worth giving
A sentence sent a day early. "Tomorrow morning you will have the import change to review, it is about an hour." That is the whole mechanism.
It is worth giving for anything that will take the recipient more than about half an hour, anything that has to be dealt with that day, and anything that will require them to stop something else.
It is not worth giving for routine small items. A team that announces everything is a team whose announcements stop being read, and the signal is lost in exactly the cases it was built for.
- Anything that will take the recipient more than about half an hour.
- Anything that has to be dealt with that day.
- Anything that will require them to stop something else.
Who has to do it
The person sending, and only the person sending, because they are the only one who knows it is coming.
This sounds obvious and is routinely got wrong, because the sender thinks of the warning as part of the handover and therefore something to write at the end of their day — at which point it arrives with the work rather than ahead of it and does nothing.
The warning has to go out before the work is ready. That is slightly awkward, because it involves predicting that something will be finished, and people are reluctant to promise. Hedged is fine: "probably tomorrow, possibly the day after" carries almost all the value.
Asking rather than telling
For anything that will take a substantial piece of somebody's day, the warning should be a question.
"Can you take the import review tomorrow?" gives the other end a chance to say no, or to say yes but not before Thursday, and both of those answers are cheap to receive the day before and expensive to receive the day after.
The instinct against asking is that it introduces a crossing. It does, and the crossing is free, because it happens during a night that was going to pass anyway while the work was being finished.
The incident case
Handing over something live is the one case where the warning is not optional.
An incident that will still be open at the handover means somebody at the other end has to clear their morning. If they learn about it when they wake, they learn about it alongside everything else and the clearing happens an hour late at best.
The practice that works is to say it as soon as it is plausible rather than when it is certain: "this may still be open when you start" sent three hours before the handover, with an update later if it closes. The false alarm costs nothing. The missing warning costs a morning.
What the receiving end can do
Not much, which is why this sits with the sender.
The one useful habit at the receiving end is to read everything that arrived before starting anything, rather than starting the plan and discovering the arrivals at eleven. That converts a disrupted day into a day planned with full information, and it is the entire argument for the reading hour that the next section is about.
It does not make the surprise cheaper. It makes it land once rather than twice.
The underlying asymmetry
Everything in this section comes back to the same thing: the sender pays seconds and the recipient pays the morning, and the two are never awake together to discuss it.
There is no mechanism that fixes that. The warning is cheap, it is not enforceable, and the person who benefits cannot ask for it in advance because they do not know the work exists. It gets done or it does not, depending entirely on whether the people sending have decided it matters.