Target audience: All users

Description: A self-help index that maps common problems to their causes and fixes. Each entry links to the page or task where the feature is documented in full.

Find your symptom in the index, then follow the link. If nothing here helps, see 7.3 Help and Support.

Possible cause 1: Display name used instead of username. CQA signs you in with your username (for example, A1234), not the display name shown in the app (for example, "Agent 1234"). A display name is rejected as invalid credentials.

Solution: Enter your username. If you do not know it, ask your CQA administrator; it is listed in the Username column of User Management.

Possible cause 2: Wrong or missing Tenant ID. The login page asks for your organization's Tenant ID as well as your username and password.

Solution: Enter the Tenant Name or ID provided by your administrator, then your credentials. See 2.2.1 Sign in.

Possible cause 3: Account locked after too many failed attempts. After 5 failed sign-in attempts within 30 minutes, the account is locked, and it does not unlock automatically after a set time.

Solution: Reset your password with Forgot Password? — a successful reset also unlocks the account. See 2.2.3 Forgot Password. If your organization uses SSO / IAMIX, reset through your identity provider instead.

Possible cause: Invitation links are valid for 72 hours, for secure onboarding.

Solution: Ask your CQA administrator to invite the same email address again in User Management. The User Management page does not have a resend action, but inviting the same email again creates a fresh 72-hour invitation link (and applies the role and username from the new invite). You do not need to contact CQA Support. See 2.2.2 First-time login and Invite a user.

Possible cause 1: Your role does not include it. For example, Usage & Limit and Policy (in the footer of the left navigation) are Admin-only; Agents do not see Test Evaluation; most Manage pages are for Admins.

Solution: Check What each role typically sees. Ask your administrator if you need access.

Possible cause 2: The page is not part of your account setup.

Solution: Contact your Customer Success Manager.

Applies to WORK › Dashboard, Analysis List, and AI Insight.

Possible cause 1: Filters are too narrow. The date range, agent, or Quality Profile filter may exclude all data.

Solution: Reset the filters or widen the date range. On a dashboard, deselect the Quality Profile to see all profiles (3.8 Dashboards).

Possible cause 2: Your access is restricted by metadata. Your account may be limited to certain metadata values (for example, one Region or Campaign).

Solution: You only see data within your access rights. Ask your administrator to check your Access Metadata in User Management.

Possible cause 3: Nothing has been analyzed yet. If no interactions were analyzed in the period, there is nothing to show.

Solution: See Incoming calls are not being analyzed.

Possible cause: The field exists but is not enabled for filtering.

Solution: An Admin goes to Manage › Data Integration › Metadata and turns on Enabled and Filter Option for the field. See Configure how a field behaves.

Possible cause 1: No Assignment Rule matches. The call's metadata does not match any active Rule-Based Assignment.

Solution:

See How rules are evaluated.

Possible cause 2: Smart Assignment found no topic match and has no fallback. If no AI Match Profile meets the confidence threshold and the rule has no Default Profiles or Fallback Profile, nothing is scored.

Solution: Add a Fallback Profile, broaden the Confidence Threshold, or improve the Topics on your profiles. See Smart Assignment profile roles.

Possible cause 3: The conversation cap is reached. When the account hits its daily, weekly, or monthly cap, further interactions are not analyzed until the period resets. API ingests return CONVERSATION_CAP_EXCEEDED.

Solution: An Admin checks Usage & Limit, in the footer of the left navigation (7.1) and contacts the CSM to change the cap. To get warned in advance, turn on Usage Approaching Plan Limit and Max Analysis Cap Reached in Alert & Notification.

Possible cause 4: Data is not arriving. Check the import job or connector — see 8.8 and 8.9.

Possible cause: Topic matching, Confidence Threshold, Default Profiles, or Fallback Profile settings on the Smart Assignment rule — or several rules matching the same call.

Solution: Review the Topics on the relevant Quality Profiles. In the Assignment Rule, check AI Match Profiles, Confidence Threshold, Default Profiles, and Fallback Profile. Tighten or broaden the threshold, or adjust topics, so the intended profiles match. Remember that every matching rule contributes its profiles (How rules are evaluated).

Possible cause: The profile already has the maximum of 10 topics.

Solution: Remove a topic before adding a new one. See Topics.

Test Evaluation is available to Supervisors and Admins.

Possible cause: Some rows in the uploaded file have invalid or unreachable audio URLs, or are missing mandatory data. On free plans, the test-evaluation limits on audio minutes and text words also apply.

Solution:

See Run a test evaluation.

Possible cause 1: The scorecard is in view mode.

Solution: Click Edit on the Interaction view to enable score changes and dispute resolution. See Resolve a dispute.

Possible cause 2: Your role cannot edit scores. Agents can raise disputes but cannot change scores; Supervisors can.

Possible cause: This is how criticality works. The effect depends on the criticality level set on the KPI.

Solution: Check the KPI's Criticality in the Quality Profile (KPI settings):

Message: "Another KPI in this quality profile already uses this title. Titles must be unique within a profile."

Possible cause: Another KPI in the same profile has the same title. The check ignores case and leading or trailing spaces, so "Greeting" and " greeting " count as the same.

Solution: Rename one of the KPIs. The same rule applies to the KPI column in Excel uploads. See KPI settings.

See 4.2 Knowledge Base.

When a KPI uses Tool Integration and the tool cannot be resolved, the analysis fails rather than scoring without verification.

See Failure behavior and Register a REST API.

Possible cause 1: Enable Notifications is off, or the event is turned off.

Solution: Check both in Manage › Configuration › Alert & Notification and click Save Changes. See Turn on notifications.

Possible cause 2: Some recipients have no email address, or an address is wrong. Alert History shows Failed or Partially Delivered.

Solution: Open Alert History, click the row's Recipients, and correct the addresses or user emails. See Review alert history.

I do not see a language toggle on the Interaction view

The detected language does not match the spoken language

The language detection model identifies the dominant language in the audio. With heavy code-switching (for example, Hindi and English mixed), it may pick the most prevalent language. This is expected for mixed-language conversations. If detection is consistently wrong for single-language calls, report it to CQA Support with the Interaction ID.

I do not see an AI Suggestion for a failed KPI

An AI Suggestion seems inaccurate or irrelevant

See AI Suggestions.

Go to WORK › Report History, set the Status filter to Failed, and read the status shown in the table to tell failure types apart.

See Test Connection vs Test Run.

Check the error.code and message in the response, and keep the request_id for support.

Full list: Error codes.

Possible cause: Callback delivery is not turned on for your account, so callback_url values are stored but not called.

Solution: Contact your Customer Success Manager. See 6.7 Webhooks / callbacks.

This is not expected; all surfaces should show the same redaction. Report it to CQA Support with the Interaction ID and the field where the difference appears.

See 4.7 PII Redaction.