The short version
- Agencies with a delivery team abroad have a specific failure: the client expects a same hour answer and the person with the answer has gone home.
- The account manager fills the gap by guessing, which is how agencies over promise without anyone deciding to.
- Separate the three things a client asks for: status, a decision, and an escalation. Only one of them genuinely needs the delivery team.
- Tell the client the coverage pattern up front. A stated pattern is professional. An implied one gets you judged against an imaginary standard.
A digital agency with account management in New York and delivery in India. The client writes at 09:00 Eastern asking whether the campaign changes went live, which is 18:30 in India and the team has finished.
The account manager has three options: say they will check and come back tomorrow, guess, or message the delivery lead at home. Over a year, agencies do a great deal of the second and third.
The shape of the problem
This is not the offshore delivery seam described in managing an offshore development team, which is about work handover. This is about a third party with their own expectations sitting on one side of the gap.
Clients do not care where your team sits. They have a question during their working day and they measure you on the answer. An agency that says "I will come back to you tomorrow" three times in a fortnight is an agency that feels slow, whatever the delivery quality.
So the constraint is fixed: answers during the client's day, from a team that is not working then.
Why the account manager guesses
Not carelessness. It is the least bad option available in the moment, and the incentives all point at it.
Saying "I do not know" repeatedly damages confidence. Messaging the delivery lead at home is uncomfortable and does not scale. Guessing is fast, usually roughly right, and the cost lands later on someone else.
The cost is real. A guess becomes a commitment the client heard, the delivery team learns about it the next morning, and now there is either an awkward correction or a piece of unplanned work. This is the mechanism by which agencies over promise without anybody deciding to over promise.
Three things a client is actually asking for
| Ask | Sounds like | Needs the delivery team? |
|---|---|---|
| Status | "Did the changes go live?" | No. This is a record question. |
| Decision | "Can you add two more variants by Friday?" | Yes, and it should wait. |
| Escalation | "The landing page is down." | Yes, now. Named on call. |
Do the count over a month of client messages. Status is the large majority on every agency account, decisions are a minority, and genuine escalations are rare. The majority category is the one that does not need a person at all, and it is the one causing the evening messages and the guessing.
Coverage patterns that work
A named on call for escalations only. One person per week in the delivery zone, with a written definition of what qualifies, reachable by phone. This is the only arrangement that should ever reach someone outside their hours, and defining it narrowly is what keeps it sustainable.
A decision window. One hour each day where the delivery lead and the account team overlap, for the decision category only. Everything in that bucket queues for it. Clients accept "I will confirm by 10am tomorrow" far better than a guess that turns out wrong.
Status without a person. The large category, and the one with no obvious answer in most agencies. The delivery team knows what went live. The account manager does not, at the moment the client asks.
This is what StandIn is built for. At the end of their day the delivery team spends about ninety seconds each on a brief: what moved, what is open, what is blocked, what is next, mostly drafted from the work that already happened. During the client's morning, the account manager asks and gets an answer from that record, in the delivery person's words, with a source under it. It never guesses, and when the answer is not in the record it says so and names who to ask, which is precisely the moment an account manager needs to hear "I do not have that" rather than a plausible sentence. The 09:00 question gets a real answer at 09:02, and nobody is messaged at home. See working across time zones.
Telling the client the pattern
Agencies hide the delivery location, or mention it once in the pitch and never again, and then get measured against an imagined standard of constant availability.
Say it plainly at the start of the engagement. Status questions answered during your business hours. Decisions confirmed by 10am the next day. Urgent issues by phone, twenty four hours, here is the number. Three lines in the kickoff document.
Clients are almost always fine with a stated pattern and frustrated by an inconsistent one. What damages the relationship is a fast answer on Tuesday and a two day wait on Thursday with no explanation of the difference.
Common Questions
Should we shift the delivery team's hours to overlap the client?
For a narrow window and with compensation, it can work. As a standing arrangement it produces turnover, and turnover on a client account costs more than a slower answer ever did.
Do we need an account manager in the delivery zone?
It helps at scale and adds a relay at small size, which means you hear about problems later and in summarised form. Fix the information flow before adding a person to carry it.
How do we stop clients from messaging delivery people directly?
Usually they do it because the account route is slow. Make the account route fast for status and the direct messages stop without a rule.
What about clients in a third time zone?
The pattern holds, the decision window gets harder to place, and it is worth being explicit that decisions take a day rather than pretending otherwise. See managing across three time zones.
90 seconds, then it's on.
Engineers publish a brief before they log off. The next timezone starts with full context, not a reconstruction of what happened while they slept.