# How To Choose Your First QA Workflow

## Objective

Choose one QA workflow that answers a useful quality question and can be checked with a small set of support conversations.

## Environment

- Your workspace has QA access and a quality owner who can manage the pilot setup.
- You have reviewed the [data and access checklist](/help/articles/qa-what-data-and-access-you-need-before-qa-setup/).
- You know which service category, Zendesk tickets, reviewers, and agents are in the first group.

## Resolution

1. **Name the decision you want QA to support.** For example, choose whether reviewers need consistent grades for one request type, whether managers need better evidence for feedback, or whether agents need a clear view of published quality results.
2. **Pick one representative category.** Use a category with enough finished, attributable conversations for a meaningful check. Open [Categories](https://wfm.zenops.cx/service/categories) to confirm the category is enabled and that the tickets you expect are classified into it.
3. **Start with a scorecard and a ticket test when the standard is new.** Agree on a few clear criteria, publish the scorecard in [QA Settings · Scorecards](https://wfm.zenops.cx/settings/service/qa-analysis/scorecards), assign it to the category, and test a selected ticket. Inspect the rationale and cited evidence in [QA Analysis](https://wfm.zenops.cx/service/quality) with a reviewer before increasing coverage.
4. **Use a bounded batch when you need to compare examples.** Start a controlled set from [QA Grade Batches](https://wfm.zenops.cx/service/quality/grade-batches), review scored and skipped results, and refine the standard or data setup where needed. A batch is separate from future automated coverage. Ticket test grades are for setup checks and cannot be published to agents.
5. **Add publication and coaching only when the review process is ready.** Decide who may adjust grades, who will approve publication, how agents can ask for a correction, and who owns follow-up actions. Publish a small set of individual agent results before extending the process.
6. **Record a pilot measure and review it.** Track whether the chosen tickets were eligible, whether evidence supported the grades, how often reviewers changed a criterion, and whether agents could understand the shared feedback. Use the findings to select the next category or widen coverage.

If your first goal is understanding the product sequence, read [How Scorecards, Evaluations, Reviews, And Coaching Fit Together](/help/articles/qa-how-scorecards-evaluations-reviews-and-coaching-fit-together/).

> Canonical page: https://zenops.cx/help/articles/qa-how-to-choose-your-first-qa-workflow/
> Audience: For administrators and managers
> Summary: Choose a small QA pilot built around one category, a clear review question, and representative tickets.