Working-Hours Agreements for Distributed Teams That Hold Up

A working-hours agreement is a short document saying when each person works, how fast each channel is answered, what counts as an exception, and what happens when someone ignores it. Without one, distributed teams do not get flexibility. They get an unbounded expectation dressed up as autonomy.

"Flexible hours turned into always available."

How flexible became always available

Flexibility is offered as a benefit and it removes a boundary without replacing it. In an office, the boundary was physical: people left, and the building emptied. Remove the building and, unless something else marks the edge of the day, the edge is set by whoever sends the latest message.

Across time zones this compounds. Someone else is always at work, so there is always a live conversation to be missing. Flexibility then means each person is free to choose which hours they work and not free to choose which hours they are asked about, and the second freedom is the one that matters.

The four parts of an agreement that holds

• Declared hours, per person, in local time. Not a company-wide window. Each person publishes their own, and they are visible to everyone.
• A response-time promise per channel. Stated in working hours, in the recipient's zone. This is the part that prevents urgency inflation.
• An exception route. One named person, one defined channel, one definition of what qualifies. If everything can be an exception, nothing is.
• A stated consequence. What happens when someone messages outside hours for a non-exception. Usually: it waits, and that is fine, and nobody has to apologise for it.

The fourth is the one that is almost always missing, and it is what turns a statement of intent into an agreement. If the consequence of an out-of-hours message is that it gets answered out of hours, the document describes nothing.

A worked agreement

Five short paragraphs. Publish it where people actually look, and pair it with the channel table in a communication plan for a team in three countries.

The senior person problem

Every working-hours agreement is tested the first time a senior person sends something at 22:00. Not because they expect an answer, usually, but because the recipient has no way to know that.

A message from your skip-level at 22:00 is never just a message. It carries an implicit question about whether replying now is what committed people do. One instance sets the norm for the quarter, whatever the document says. The two fixes are behavioural: senior people either observe the hours themselves, or they label the message explicitly ("no reply needed before Monday") every single time. Doing it occasionally is worse than not doing it, because it teaches people that the absence of the label is meaningful. More on this in the 22:00 message.

Hours you set are hours that get covered

Agreements erode for one practical reason. Someone genuinely needs an answer outside your hours, and the only way to get it is you. You know that, so you check. The document said off; the situation said otherwise; the situation wins.

An agreement holds when the hours you declare are hours your work is still answerable in. That is what StandIn does: ninety seconds at the end of your day confirming a brief, mostly drafted from the work you already did. Whenever you are off, your StandIn is on, answering questions from that brief in your words, with a source under every answer, clearly labelled, and never guessing. The colleague in the other zone gets their answer during their working day. You are genuinely off, rather than nominally off.

The hours you set become hours your StandIn covers, which is what makes them credible. Log off like you mean it. See working across time zones.

Common Questions

Should working hours be company-wide or per person?

Per person, published in local time. A company-wide window either excludes somebody's zone or is so wide it means nothing. What should be company-wide is the response-time promise and the definition of an exception.

How do we handle people who prefer to work evenings?

Let them, and separate sending from expecting. Someone choosing to work at 21:00 is fine. Someone at 21:00 expecting replies from colleagues who declared 09:00 to 17:00 is not. Scheduled send exists precisely for this.

What if a client or customer expects out-of-hours responses?

Then that is a role with a rota and compensation, and it should be small, named, and deliberate. What you want to avoid is a client expectation quietly becoming everyone's expectation because nobody set the boundary at the start of the relationship.

Does a written agreement actually change behaviour?

Only in combination with two other things: senior people visibly observing it, and the questions still getting answered inside the declared hours. A document alone changes nothing. A document plus those two changes the team within a month.