BPO quality parameters are the criteria a QA analyst checks on each customer interaction to decide whether an agent handled it to the client’s standard. The usual seven are opening, verification, understanding the issue, resolution accuracy, process compliance, communication and closing. Each parameter becomes a line on the QA scorecard, and the most serious ones are marked fatal.
In short
- Quality parameters in a BPO are the checks made on one interaction; KPIs such as AHT or CSAT measure results across many interactions.
- Most BPO scorecards group their parameters into seven families: opening, verification, understanding, resolution, process compliance, communication and closing.
- Parameters are split into fatal ones, which fail the interaction outright, and non-fatal ones, which cost points.
- A hygiene audit checks the basic process steps around the conversation, such as tagging, notes and disposition, and usually runs next to the main QA audit.
- The client decides the parameters and weights; the BPO’s job is to score them consistently and prove it.
What are quality parameters in BPO?
Quality parameters in BPO are the specific behaviors and steps a QA analyst checks when auditing one chat, email or call, such as whether the agent verified the customer, gave the correct answer and followed the client’s process. Each parameter is scored per interaction, and together they make up the QA score.
Many guides that rank for this term list KPIs instead: average handle time, CSAT, first contact resolution. Those matter, but they are outcomes measured across thousands of contacts. A parameter is what the analyst looks for inside a single conversation. The difference decides what you can coach.
| Quality parameter | KPI | |
|---|---|---|
| Measured on | One interaction | Many interactions over a period |
| Example | “Agent confirmed the order number before sharing delivery details” | First contact resolution of 78% this month |
| Who sets it | The client, in the scorecard | The client and the BPO, in the contract |
| What it tells you | What this agent did well or badly, and why | Whether the account is on target |
| Used for | Coaching and calibration | Reporting and SLA reviews |
The QA score links the two: it is the average result of the parameters across evaluated interactions, and it is usually one of the KPIs in the contract. For the full list of BPO KPIs with formulas, see our guide to KPI meaning in BPO. Teams that score many clients’ parameters at once often run them on a BPO QA platform rather than in one spreadsheet per client.
What are the 7 quality parameters in BPO?
The 7 quality parameters most BPO scorecards use are opening, verification, understanding the issue, resolution accuracy, process compliance, communication and closing. Names vary by client, and each family holds two to five detailed checks. Together they cover what the agent said, what the agent did and whether the customer got the right outcome.
Here are the seven, with what a QA analyst checks and an example of a miss on a chat or email account.
| # | Parameter | What the QA analyst checks | Example of a miss (chat or email) |
|---|---|---|---|
| 1 | Opening | Greeting in the client’s format, agent name, acknowledging the customer’s message | The agent replies to an email without addressing the question in the first lines |
| 2 | Verification | Identity or account checks the client requires before sharing or changing data | The agent shares an order’s delivery address before confirming the customer’s email |
| 3 | Understanding the issue | Reading the full message, asking the right clarifying questions, not making the customer repeat themselves | The agent asks for an order number the customer already gave |
| 4 | Resolution accuracy | Correct and complete answer, per the client’s knowledge base and policies | The agent promises a refund that the policy does not allow |
| 5 | Process compliance | Required steps, tools and macros used correctly, escalations raised when needed | A chargeback request is answered instead of escalated to the payments team |
| 6 | Communication | Tone, empathy, grammar, clarity, brand voice | A frustrated customer gets a copied macro with no acknowledgement of the problem |
| 7 | Closing | Confirming the issue is solved, next steps, any required closing statement | The chat ends without telling the customer when the refund will arrive |
Some clients add an eighth family for documentation and tagging. Others fold it into process compliance or audit it separately in a hygiene audit (see below).
The weight of each parameter is the client’s choice. On a payments account, verification and process compliance carry most of the points; on an e-commerce account, resolution accuracy and communication often do. Our guide on how to weight QA scorecard criteria shows how to decide.
What are fatal and non-fatal parameters in BPO QA?
Fatal parameters are errors serious enough to fail the whole interaction, usually scoring it zero, such as skipping a mandatory verification, giving wrong financial or legal information, or a data protection breach. Non-fatal parameters cost points but do not fail the interaction, such as a missing greeting or a spelling mistake. Most scorecards report both separately.
Fatal errors are also called critical errors or auto-fails. Keep them few and unambiguous; if analysts argue about whether something was fatal, the definition is not tight enough. Our explanation of what an auto-fail is in QA covers how to write them.
Many BPOs split fatal parameters into three kinds:
- Customer fatal: the customer is harmed or misled, for example a wrong refund amount or a promise the client cannot keep.
- Business fatal: the client loses money or reputation, for example an unauthorized discount or a rude reply that gets shared.
- Compliance fatal: a legal or regulatory rule is broken, for example sharing account data without verification.
Report the fatal error rate next to the average QA score. An account with a 90% average and 4% of interactions failing on a fatal parameter is in worse shape than its average suggests.
What is a hygiene audit in BPO?
A hygiene audit in BPO is a check of the basic process steps around an interaction rather than the conversation itself: correct tagging or disposition, case notes, ticket status, CRM fields updated and the right macro or template used. Teams use the term differently, so agree the exact list with the client before you report on it.
Hygiene parameters look small, but they feed everything the client sees. Wrong tags break the reporting on contact reasons, missing notes force the next agent to ask the customer again, and a ticket closed in the wrong status hides a backlog.
A typical hygiene checklist for a chat and email account:
- Contact reason tag matches what the customer actually asked about.
- Case note states the issue, the action taken and any promise made.
- Ticket status and priority are correct, and follow-ups have a due date.
- Customer and order fields in the CRM or helpdesk are updated.
- Escalations went to the right queue with the required information.
Hygiene audits often run on a larger sample than the full QA audit, because each check is quick. When conversations are scored automatically, the hygiene checks can sit on the same scorecard as the quality parameters.
Are call quality parameters different from chat and email parameters?
Call quality parameters follow the same seven families as chat and email: opening, verification, understanding, resolution, process, communication and closing. What changes is how communication is judged. Many contact centers add voice-specific checks to calls, such as hold procedure, silence and talking over the customer, while written channels add grammar, formatting and response structure.
For written channels, the communication parameter usually breaks down into:
- Chat: response time between messages, short and clear messages, no unexplained pauses, checking the customer is still there before closing.
- Email: a subject and opening that address the question, one complete answer instead of several partial ones, correct grammar and formatting, a clear next step.
Keep one scorecard per client where you can, with channel-specific lines only where the channel really changes what good looks like. Separate scorecards per channel make it hard to compare agents who work several channels. If your team also needs to improve the scores those parameters produce, our guide on how to improve quality score in BPO covers the coaching side.
How do you turn quality parameters into a BPO QA scorecard?
You turn quality parameters into a scorecard by writing each one as a yes or no question an analyst can answer from the conversation, giving each a weight, marking the fatal ones, and testing the scorecard on real interactions with two analysts before using it for agent results. The client signs off the final version.
A worked example: here is one email audited on a seven-parameter scorecard for an e-commerce client. The weights are illustrative.
| Parameter | Weight | Result on this email | Points |
|---|---|---|---|
| Opening | 5 | Addressed the question in the first line | 5 |
| Verification | Fatal | Confirmed order number and email before sharing details | Pass |
| Understanding the issue | 15 | Asked for a photo the customer had already attached | 5 |
| Resolution accuracy | 35 | Correct replacement offered per policy | 35 |
| Process compliance | 20 | Replacement logged, but no internal note | 10 |
| Communication | 15 | Clear and polite, no acknowledgement of the delay | 10 |
| Closing | 10 | Told the customer when the replacement ships | 10 |
| Total | 100 | 75% |
The QA comment the agent receives should name the evidence, not just the score: “You offered the right replacement and gave a clear ship date. Two points to work on: the customer had already attached the photo in their first email, so asking again added a day; and add an internal note when you log a replacement, so the next agent sees what was promised.”
Before the scorecard goes live, have two analysts score the same 20 to 30 interactions and compare. Where they disagree, the parameter wording is usually the problem. Our guide to running QA calibration sessions explains the routine, and the quality feedback examples show how to phrase comments agents act on.
How should the parameter set change with team size?
The parameter set should get more detailed as the account grows, but the seven families stay the same. A small account needs one short scorecard and a manual sample; a large multi-client BPO needs one scorecard per client, regular calibration across sites and scoring that covers far more than a sample. Pick the setup that matches your size.
| Account size | Scorecard | How interactions are selected | Calibration | Tooling |
|---|---|---|---|---|
| Small (one client, under about 15 agents) | 7 families, 10 to 15 checks, 2 or 3 fatal parameters | A fixed number per agent per week, picked at random | Monthly with the client | A shared sheet or the helpdesk’s own QA feature |
| Mid-size (several queues or clients, 15 to 100 agents) | One scorecard per client, hygiene checks added | Random sample plus targeted reviews of low CSAT and escalations | Every two weeks across analysts | A QA tool with per-client scorecards and dispute handling |
| Large (many clients, 100 agents or more, several sites) | One scorecard per client and channel where needed, versioned | Automated scoring of every interaction, analysts review flags and disputes | Weekly, across sites and with each client | Automated QA with per-client reporting and evidence for every score |
These are starting points, not rules. The client contract can require a fixed sample size, and regulated clients may require extra fatal parameters whatever your size.
How a BPO scores every client against its own parameters
When a BPO runs many clients, each with its own parameters, the hard part is consistency: every analyst has to apply many different definitions of good, one per client, without mixing them up. Spreadsheets break first, because nothing stops an analyst from using the wrong client’s standard.
EverHelp, a BPO that has delivered more than 100 client projects, ran quality control on Google Sheets until that point. After moving to Kaizo, which scores each conversation against each project’s own criteria, its QA coverage rose 270% and tickets handled per hour rose 280% between 2023 and 2024, and coaching prep dropped 75% across sixteen domains. The EverHelp customer story explains what changed.
Score every client account against its own parameters. Kaizo scores 100% of your support conversations against each client’s scorecard, with the evidence behind every result. Book a demo.
When a detailed parameter set is not for you
A detailed parameter set is more than you need when the account is new, small or has no agreed quality standard. Five clear checks that the client and your analysts agree on beat thirty that nobody applies the same way. And no tool fixes parameters that were never written down.
Be honest about where you are:
- No signed-off scorecard yet: agree the parameters and fatal errors with the client first, then think about tools.
- A handful of agents on one client: a manual sample on a short scorecard, reviewed with the client every month, is usually enough.
- Mostly calls: the parameter families on this page still apply, but check which channels a QA tool scores before you buy, and test it on your own mix of channels.
For the governance side of outsourced quality, including who scores and how client and vendor share one standard, read our guide to QA meaning in BPO. When the parameter set is stable and the sample is the bottleneck, see how Kaizo for BPOs scores each client account.
Frequently asked questions
What is the difference between QA parameters and quality parameters in BPO?
They usually mean the same thing: the criteria on the QA scorecard used to audit each interaction. Some teams say QA parameters for the scorecard lines and quality parameters for the wider set that includes hygiene checks, so confirm the meaning with your client.
How many quality parameters should a BPO scorecard have?
Most scorecards work well with the seven parameter families and 10 to 20 detailed checks in total. More checks make audits slower and calibration harder. Fatal parameters should stay few, usually two to five, and each must be unambiguous.
Who decides the quality parameters in a BPO?
The client decides, usually with the BPO’s quality team. The client knows its policies, compliance rules and brand voice; the BPO’s QA team turns them into checks an analyst can score consistently. Both sides sign off the scorecard and its version.
What is a good QA score in a BPO?
There is no universal good QA score, because it depends on the parameters, their weights and how strict the fatal errors are. Many contracts set the target after a baseline period on the agreed scorecard. Always read the score next to the fatal error rate.
What is a fatal error in a BPO call or chat?
A fatal error is a mistake that fails the whole interaction, whatever the other parameters scored. Common examples are skipping mandatory verification, sharing data with the wrong person, giving wrong financial information, or abusive language. The client’s scorecard lists exactly which errors are fatal.
How often should BPO quality parameters be reviewed?
Review them whenever the client changes a policy, product or process, and at least at each quarterly business review. Parameters that analysts keep disagreeing on in calibration should be rewritten straight away rather than at the next review.
