Hiring guides
Is offshore customer support actually any good?
The honest answer: it depends almost entirely on which of three arrangements you picked, and most people who had a bad experience picked the same one. Here is what separates the outcomes.
The Adarna Team
Editors
You have heard the horror stories, and some of them are true. Scripted replies that do not read the question. An agent who vanishes mid-conversation. A team that changes every quarter so nobody ever learns the product. If that is your mental picture of offshore support, it is not unfair. It is just describing one specific arrangement, which people then generalise into a verdict on an entire country's workforce. The useful question is narrower: what actually separates the good outcomes from the bad ones?
The bad experiences cluster
Ask around and the disasters are not randomly distributed. They concentrate in two arrangements, and both fail for structural reasons rather than because of who was answering the tickets.
The shared-agent model. You buy seats from an outsourcing firm. Your customers become tickets in a queue that also holds three other companies' tickets. The agent handling yours at 2pm handles someone else's at 4pm. Nobody learns your product, because learning your product is not what they are paid for, and the person who did learn it has been rotated to another account. The quality problem here is not the agent. It is that you bought capacity when you needed a colleague.
The lottery model. You post on a marketplace, get two hundred applications, cannot assess two hundred applications, so you pick on price and a plausible profile. You have not tested written English under pressure or judgement on an ambiguous ticket, because there was no mechanism to. Sometimes it works brilliantly. Often it does not, and the failure feels like proof about offshore hiring when it is really proof about hiring without assessment. You would get the same distribution hiring in Manchester that way.
Almost nobody who says "we tried offshore and it did not work" means they tried a vetted, dedicated person and it did not work. They mean they tried a shared queue, or they tried a lottery.
What the third arrangement changes
One dedicated person, assessed before you meet them, working your hours, in your helpdesk, on your team. The variable that moved is not geography. It is that they are yours, and that someone tested them before you bet on them.
| Shared BPO seat | Marketplace hire | Vetted, dedicated | |
|---|---|---|---|
| Learns your product | No incentive to | Maybe | Yes, it is their job |
| Assessed before you commit | Not by you | No | Yes |
| Continuity | Rotates | Unpredictable | Stays |
| Who they work for | The BPO | Themselves | You |
| Failure mode | Quality drift | Coin flip | A person leaves, you replace them |
Why the Philippines specifically
Not because it is cheap. Cheap is available everywhere and buys the same thing everywhere, which is churn. The reasons that hold up are narrower: English is an official language and the medium of instruction through university, so written fluency is common rather than exceptional. There is a deep, established customer-service profession with real career paths, which means you are hiring from a trade, not from a pool of people trying something. And the cultural norms around service, which is the part nobody can quite quantify, tend to run warm rather than transactional.
The honest counterweight is the timezone. UK hours are an evening and night shift in Manila. Anyone telling you that is costless is selling something. It is a real ask, it deserves a night differential in the offer, and it is the thing most likely to wear a person down over two years if you are careless about it.
When the answer is genuinely no
There are cases where we would tell you not to do this, and it is worth being specific rather than pretending our answer fits everyone.
- You need someone in the room. If the role is really office-manager-plus-support, or it involves physical goods, hire locally.
- Regulated or licensed advice. Financial, legal or medical guidance with a compliance perimeter. Do not improvise this offshore.
- Your customers expect a British voice on a phone. If that is genuinely your market's expectation, respect it. Accent coaching is not a strategy and everyone can tell.
- You will not actually manage them. This is the real one. A dedicated person needs onboarding, a weekly conversation and someone to escalate to. If you want to hand the problem to a supplier and never think about it, buy a BPO seat and accept what that buys. A dedicated hire you ignore will fail, and it will not be their fault.
What to ask anyone selling you this
Including us. If the answers are vague, the quality is a coin flip regardless of the logo on the deck.
- Is this person dedicated to me, or shared? If shared, say so plainly.
- What exactly did you test, and can I see the output rather than a score?
- Who employs them, and what do they actually take home?
- What happens when it does not work out, and who pays for that?
- Can I speak to the person before I commit, not just read a profile?
We answer those with a two-stage assessment you can inspect: written English, judgement, a recorded live interview and a chat simulation, scored on content, with a human making the decision. You see the recordings and the scores before you meet anyone, then you interview the three and choose. The standard being scored against is public, in what good looks like in first-line support.
You should not take our word for the quality. That is the entire point of showing you the evidence instead.