Tier-one helpdesk staffed and run for you

What this actually solves

Two costs sit behind almost every enquiry for this service. The first is churn you cannot see. When answers arrive slowly, customers do not complain — they leave, quietly, and you pay to win a replacement without ever being told why the first one went. The damage shows up in a renewal report months later with no obvious cause attached, and by then the public reviews already say the product is good and the support is not.

The second is founder time. Two developers working the support inbox every day are not building anything. They are also, in most cases, bad at it — not through indifference but because support is a discipline with its own craft, and constant interruption destroys the deep work they were hired for. Handing the queue to people who do it properly buys back the expensive hours and usually improves the answers at the same time.

What is being bought here is not a person who replies. It is a system — a written service level agreement, a macro library, a knowledge base, an escalation route, a roster, and a monthly report. That distinction matters commercially too. Anyone can be cheaper than you next year. A desk that reduces the number of tickets a product generates is much harder to replace.

How the work runs

1. Discovery on real numbers, not estimates. Ticket volume for the last three months by channel and by day of week, the twenty most common questions, where your customers are and when they write, who answers now and how long it takes them, which systems we need access to, and the authority limits your agents will hold. That last item is the one people skip and it causes the most damage later. Three to five days.
2. The playbook. Coverage window and holidays with notice periods. First response and resolution targets split by urgency. The escalation matrix — which situation goes to whom, within what time, by which channel, and what the customer is told while they wait. Agent authority in plain figures. Three to five days.
3. Macros and knowledge base, written before the first ticket. Between fifteen and twenty-five saved replies covering the questions your data says actually arrive, including several that deliver bad news, because refusing a refund or explaining that a feature does not exist is where new agents freeze and write something you would not want sent. Articles are titled the way a customer would search, not the way your team refers to things internally. One to two weeks.
4. A two-week pilot. Half the queue ours, half yours, with a thirty-minute review every evening. This is deliberately the period when mistakes should surface, while your team is still watching, and it is also when we measure real average handling time. Only your own measured numbers are worth pricing against — handling time follows your product and your customers, and a price set from a borrowed range has to be renegotiated the moment the real work is known. Two weeks.
5. Full handover, roster and daily rhythm. Shift start begins with reading the previous handover note and checking open urgent tickets. Shift end is spent writing the next one — what is open, what is waiting on someone, who is angry, what needs looking at first tomorrow. The day that note does not get written is the day a customer is left hanging. Ongoing.
6. Monthly report. First response and resolution times as median and ninetieth percentile, satisfaction score, volume and trend, and the five causes generating the most tickets with a recommendation against each. Monthly, roughly a day to produce.

What moves the price

The rate is $800–$4,000/agent/mo and it is ongoing. Four things set the position.

Coverage hours. Weekday business hours is the cheapest arrangement available and round-the-clock is the most expensive by a wide margin, for the headcount reason above. Overnight shifts also carry a premium for the people working them, and that premium belongs in the price rather than being absorbed, because unpaid night work produces resignations in the first quarter.

Difficulty and language. General e-commerce questions and technical product support are different skill levels and cannot be staffed identically. Every additional language narrows the hiring pool and moves the rate with it.

How much responsibility comes with the ticket. Drafting replies for your team to send is one price. Owning a ticket to closure, with refund and discount authority and a satisfaction score attached to the outcome, is another.

What usually goes wrong

Round-the-clock coverage is promised on a small team. It works for a while on goodwill and overtime, then quality falls, then people leave, and you are recruiting under pressure with a live queue. The 168-hour week does not negotiate. If someone quotes twenty-four hour cover without discussing headcount, they have not done the calculation and you will discover the consequence, not them.

A per-ticket price is agreed before handling time is measured. You quote a monthly count from memory; the real figure includes everything that never reached a formal count, and some of those are the long investigative tickets rather than the kind a macro closes. Two weeks of measurement before any per-ticket commitment, always — and a written review if volume moves materially.

Agent authority is never defined. Without a refund ceiling and a clear list of what can be decided at the desk, every routine case becomes an escalation. The customer waits for a decision that could have been made on the spot, your team is interrupted anyway, and the desk becomes a slower version of what you had before. One line in the agreement prevents it.