Quality Assurance Outsourcing
An independent quality programme for a support team that stays yours, including verification of a third-party supplier's reported numbers.
Page guide
On this page
An independent read on a team we do not run
This is quality assurance as a standalone engagement. We design and run the quality programme for a support operation that stays entirely yours - your in-house team, or a team another supplier delivers.
Most clients arrive here for one of three reasons. They suspect a quality problem and cannot evidence it. They have quality scores that are consistently high while customer satisfaction is not. Or they have a supplier reporting their own performance and no independent way to check it.
All three are the same underlying problem: no trustworthy measurement. That is what this service provides, and it does not require you to change who delivers the support.
Why independence changes the answer
A team scoring its own work has an unavoidable conflict, and it rarely shows up as dishonesty. It shows up as a scorecard that drifts towards what the team does well, criteria that soften where they were repeatedly failed, and calibration sessions that converge on generosity because nobody wants to be the harsh reviewer.
The same applies to a supplier reporting on itself. Their numbers may be accurate and you have no way to know, which means you cannot act on them with confidence in either direction.
We have no stake in the result. Where we also deliver support for you, the QA engagement is scoped and reported separately for the same reason - see how quality works on accounts we deliver.
What the engagement includes
- Scorecard design or review - building criteria weighted on resolution, or diagnosing why your existing scorecard produces flattering numbers.
- Sampling - random for the baseline, targeted at high-risk contact types, triggered by signals such as repeat contact or a negative satisfaction score.
- Interaction review - calls, chats, emails and tickets scored by trained reviewers.
- Calibration - reviewer-to-reviewer, and with your team where you want it.
- Coaching recommendations - the specific behaviour to change, delivered to your team leaders rather than to your agents.
- Root-cause analysis - separating agent performance from knowledge, policy, process and system causes.
- Reporting and trend analysis - including inter-reviewer variance as a health check on the programme itself.
- Supplier performance verification - where you want an independent check on a third party's reported numbers.
Where we stop
We review and report. We do not manage your agents, deliver coaching to them directly, or make employment decisions. Coaching lands with the team leader who owns the agent, because feedback from someone with no authority over your development is easy to dismiss and unfair to receive.
That boundary also protects the independence. A reviewer who also manages the person being reviewed is back in the conflict this service exists to remove.
The findings clients least expect
Two patterns recur often enough to name.
The first is that the scorecard was the problem. Criteria measuring script compliance rather than resolution produce agents who follow every step and leave customers needing to contact again. The score is real and useless. Rewriting the scorecard changes the numbers more than any coaching would.
The second is that most failures are not agent failures. When several agents fail the same way on the same contact type, the cause is usually a missing or wrong knowledge article, an ambiguous policy, a process step that does not work as documented, or a system that makes the correct action hard. Coaching around those is the expensive way to hide them.
We report both plainly, including when the finding is uncomfortable for whoever commissioned the review.
How an engagement runs
- Scope. Which channels and contact types, sample volume, and what decisions the findings need to support.
- Baseline. An initial review against your existing scorecard, or ours where you have none, to establish where you actually are.
- Scorecard work. Design or revision, agreed with you before it becomes the standard.
- Steady state. Sampling, scoring, calibration and reporting on an agreed cycle.
- Review. Periodic reassessment of whether the criteria still match your product, policy and priorities. A scorecard that never changes is measuring a business that has moved on.
Coverage, languages and systems
Review work is asynchronous, so it does not need to match your operating hours - though calibration sessions and reporting reviews are scheduled in overlapping time.
Linguistic review requires reviewers who speak the language being reviewed, which an English-language scorecard cannot substitute for. From India we provide review capability in Hindi, Hindi-English bilingual, English, Bengali, Nepali and Punjabi, with further Indian languages on request - see Hindi customer support for what linguistic review actually examines.
We work in your quality, help desk and recording platforms. Where you have no quality tooling, reviews can run from recordings and transcripts directly. See technology. Access is least-privilege and scoped to review only - see security and data protection.
What you receive
Quality score by agent, team and criterion, with movement over time. Inter-reviewer variance. Root-cause findings with owners. Coaching recommendations per agent for your team leaders. Where relevant, an independent verification of a supplier's reported figures against your own sampled evidence.
Figures published elsewhere on this site are historical or representative and campaign-dependent, and we do not publish client quality results.
Getting started, and what it costs
Priced per dedicated reviewer per month with a five-person minimum - published rates and what moves them. Sample volume and channel mix are the main variables.
Bring your current scorecard if you have one, your satisfaction and repeat-contact figures, and the specific suspicion that brought you here. A well-defined question produces a much more useful baseline than a general audit. Talk to us.
Frequently asked questions
Yes - that is what this service is. Your in-house team, or a team another supplier delivers. Nothing about who delivers the support needs to change.
A team scoring its own work has an unavoidable conflict, and it rarely looks like dishonesty. It looks like a scorecard drifting towards what the team does well, criteria softening where they were repeatedly failed, and calibration converging on generosity. The same applies to a supplier reporting on itself.
Yes. We sample independently and compare our findings against their reported figures, so you can act on the numbers with confidence in either direction.
No. We produce coaching recommendations for your team leaders, who own the agent relationship. Feedback from someone with no authority over your development is easy to dismiss and unfair to receive - and a reviewer who also manages the reviewed person is back in the conflict this service removes.
Usually a scorecard measuring script compliance rather than resolution. Agents follow every step, score in the nineties and leave customers needing to contact again. Rewriting the scorecard changes the numbers more than any coaching would.
Yes, and an English-language scorecard cannot substitute. Linguistic review examines whether respectful register held under pressure and whether an explanation was genuinely clear in that language - things an English scorecard cannot see.