Offshore quality is decided at the handover

When an Australian agency or utility complains about offshore support, the complaint usually arrives framed as geography. The people are too far away, the accent is hard to follow, the time difference slows things down. I lived in Manila for five and a half years and have run offshore centres for most of my career since, and I have rarely found geography to be the cause. Manila is two hours behind Brisbane. Pune is four and a half. Neither gap explains why a user rings back three times about the same fault.

The cause is almost always in the design of the work: what the offshore team is measured on, what it is permitted to fix, how much context arrives with each ticket and where the right to make a privileged change sits. Those are decisions made in Australia, usually by people who have not spent a week on the floor where the work is done.

A ticket closed in Manila and reopened in Brisbane

In 2022 Torvelle's desks in Manila and Pune handled about 2.1 million contacts a year. Their reported first-contact resolution was 74 per cent, which looked respectable. Yet 18 per cent of resolved tickets were reopened within ten days, and we were paying service credits that had reached A$7.2 million a year across our contracts.

Listening to calls explained it. Agents were measured on tickets closed and on keeping handling time under nine minutes. They were not allowed to reset a multi-factor token or reinstall a standard application, both of which had been reserved for second-line engineers in Australia. So a large share of calls ended with sensible advice, a closed ticket and a user who would be back. The agents were doing what the measures rewarded. Their managers knew the tickets would return, and so did ours. Nobody had been asked to fix the design.

The knowledge base made it worse. Most articles had been written by onshore engineers for other onshore engineers, assuming context an agent in Cebu could not have: which agency used which version of a case-management system, which building's printers had been replaced, which business unit was mid-migration. Handover notes between shifts were thin because nobody was measured on them. In Pune, where the working day ends while Brisbane is still at its desks, an unfinished ticket often passed to an Australian engineer with a single line of context and no record of what had already been tried.

We removed closure quotas, measured every team on fixes still holding after ten days, gave first-line agents controlled tools for the commonest repairs and moved second-line engineers into the same pods. Offshore team leaders began writing and owning knowledge articles. Handling time rose 21 per cent and we hired 170 people. Reopened tickets fell to 6 per cent, user ratings rose from 3.1 to 4.2 out of five, and Manila attrition fell from 41 to 23 per cent. The same people, given a different design, produced a different result. That is the most important thing I know about offshore delivery.

Move the judgement offshore and keep the privilege home

The opposite error is now common. After several serious breaches in Australia, some clients and suppliers have concluded that offshore delivery itself is the risk, and that the safe answer is to bring work home. I understand the instinct, and for some contracts, where data must stay in Australia and staff must be cleared, it is already the rule. As a general answer, it confuses two different things: where judgement is exercised and where privilege is held.

Since 2024 Torvelle has separated them for government and critical-infrastructure clients. Offshore engineers diagnose faults, analyse logs, prepare changes and test them outside production. The privileged step, the change to a live system, is made only by a cleared engineer in Australia, who requests access for that task, works from a managed workstation and is recorded throughout. The offshore engineer who prepared the change often watches it being made. The work is still shared; the standing keys are not.

This cost us. Ninety-six onshore roles added A$13.5 million a year. Seventy roles in Pune were redesigned, and eleven engineers resigned within six months, some of them among our best, because they read the change as distrust. They were not entirely wrong to feel it, and I handled the announcement badly. But in February 2025, when a phishing campaign captured the credentials of three of our engineers, none of the accounts could reach a client system. Removing standing privilege did more for our clients than moving any number of desks.

What the cyber strategy asks of suppliers

Australia's cyber security strategy for 2023 to 2030 treats protection of critical infrastructure and of the technology supply chain as work that government and industry share. For a supplier, the practical effect arrives through clients. Water, energy and health businesses that are responsible entities under the Security of Critical Infrastructure Act must maintain risk-management programmes covering cyber, personnel, supply-chain and physical hazards. A managed-service provider with administrator access is one of the largest supply-chain hazards they have, and their risk teams now ask us precise questions: who can reach our systems, from which country, with what approval and with what record.

I welcome those questions, because they move the conversation from geography to design. A supplier that can show its privileged sessions by person, place and approval has a better answer than one that can only promise everything is done in Australia. Restricting administrative privileges is one of the Essential Eight mitigation strategies, and it is far easier to evidence than the location of a desk.

For directors of a company that buys or sells offshore services, I would suggest four questions. What share of privileged sessions on critical systems were performed from outside Australia last quarter, and by whom? What is the reopen rate by delivery site, not only the closure rate? What is attrition by site, and how long does a new agent take to reach full competence? And who wrote the knowledge articles the offshore teams rely on? A management team that can answer those questions has designed its offshore delivery. One that can describe only its savings has merely relocated it.