Analytics Policies

Analytics Policies control which post-call analytics are applied to interactions. Admins can build policies that trigger based on interaction and agent metadata. When a policy matches, the assigned analytics are applied.

How Analytics Policies Work

Analytics Policies use a Filter and Scope approach to determine which interactions are processed and for which analytics.

  • Filters: Criteria like Skills, Teams, or Agent Attributes define which interactions match the policy.

  • Policy Scope: Once an interaction matches the filters, the scope defines the transcription language, redaction rules, and which analytics services (like IA and Topic AI) are applied.

The "First Match" Logic

Policies are processed in a top-down hierarchy. When an interaction occurs, CXone checks the list of policies starting at Priority 1:

  1. Match Found: The system applies that policy’s scope and stops searching. No other policies—even those with more services—will be applied to that interaction.

  2. No Match Found: The system moves to the next policy in the list.

  3. No Match at End of List: If an interaction doesn't match any policy, it will not be transcribed or analyzed.

There is no longer a "default tenant language." You must ensure every interaction you want analyzed is covered by at least one policy.

Understanding Filter Logic

To avoid configuration errors, ensure the following when setting up filters:

  • OR Logic (Inside a Filter Type): If you select "Sales" and "Support" in the Skills filter, the policy applies to interactions from either skill.

  • AND Logic (Across Filter Types): If you select "Sales" (Skill) and "Can Be Analyzed" (Agent Attribute), the policy only applies to interactions that meet both criteria.

Supported Languages

Following Languages are supported:

Languages
  • English
  • Portuguese
  • Spanish
  • German
  • Dutch
  • French
  • Japanese
  • Italian
  • Welsh
  • Chinese Mandarin- Simplified
  • Korean
  • Swedish
  • Finnish
  • Norwegian
  • Danish
  • Hungarian
  • Filipino

  • Tagalog

  • Thai

  • Vietnamese

  • Greek

  • Cantonese

  • Russian

  • Croatian

  • Polish

  • Indonesian

  • Malay

  • Turkish

  • Czech

  • Ukranian

  • Galician

  • Romanian

  • Catalan

  • Bosnian

Key Actions Available

  • Create a New Policy: Use the New button at the top-right corner.

  • Prioritize Policies: Use the Prioritize button to change policy priority.

  • Edit/Delete/Activate/Deactivate a policy: Click the stacked dots icon in the Action column next to the policy and choose the relevant option.

Available Policies Table

This table lists all Analytics Policies configured for your tenant.

The table includes:

Column Name

Description
PRIORITY Order in which policies are evaluated. Policies at the top (for example, P1) are evaluated first.
POLICY NAME Name of the policy.
DESCRIPTION Optional description of the policy.
LANGUAGE Analytics language applied when the policy matches (for example, English, French, Dutch). Also defines the language used by post-call transcription on voice calls.
CREATED DATE Date and time when the policy was created.
LAST MODIFIED Date and time when the policy was last updated.
STATUS Current state of the policy, such as Active or Inactive.
ACTION Opens a menu to edit, delete, activate, or deactivate.

Create an Analytics Policy

To create a new analytics policy:

  1. Click the app selector icon of app selector and select Admin> Analytics Services > Analytics Policies from the left navigation panel.

  2. On the Analytics Policies page, click + New Policy. This opens the New Analytics Policy page.

  3. In the General Definitions section:

    • Policy Name: Enter a unique name for the policy.

      • Maximum 64 characters. A character counter below the field shows how many characters you have used out of the allowed limit. You cannot enter more than 64 characters.

      • Special characters like | : * / ! \ + < > ? # & , % " are not allowed: If the value contains disallowed characters, a validation message appears, and you must correct the value before saving.

    • Language: Select the language that will be used for the post-call transcription and analytics.

    • Description: (Optional) Enter a description of what this policy covers.

      • Maximum 200 characters. A character counter below the field shows how many characters you have used out of the allowed limit. You cannot enter more than 200 characters.

      • All special characters are allowed.

  4. In the Redaction Profile section, select a profile from the drop-down list to apply specific data masking rules to the interactions caught by this policy.

  5. In the Policy Scope section, select the analytics modules you want to run for these interactions:

    • Interaction Analytics: Enables category and sentiment analysis.

    • Quality Management Analytics: Enables QM-specific metrics and evaluation capabilities.

    • Topic AI: Triggers automated topic identification.

    • NICE AI CSAT / Sales Effectiveness: Enables specific NiCE AI metrics.

    Note: Available modules depend on your tenant's licenses.

  6. In the Filters section, configure the criteria used to identify interactions for this policy. You can select multiple types:

    1. ACD Skills: Select specific skills from the list (these are the skills defined in the admin tab).

    2. Agent Attributes: Filter by attributes like "Can Be Analyzed" or "Can Be Evaluated."

    3. Business Data: Define rules based on custom business data fields (e.g., "Order Value is greater than 100"). You can add up to 10 business data fields per policy.

    4. Direction: Choose Inbound, Outbound, or both.

    5. Teams / Employees: Select specific organizational units or individual agents.

  7. When you finish configuring the policy, click Save. If any field-level validations fail, inline messages appear next to the affected fields, and the policy is not saved until those issues are resolved.

  8. If the policy is created successfully, you’ll see a confirmation message, and the new policy appears in the grid.

Manage and Prioritize Policies

Policies are evaluated in a top-down hierarchy. The first policy that matches an interaction's criteria is the only one applied. To prioritize a policy:

  1. On the Analytics Policies page, click Prioritize to open the Prioritize Analytics Policies window.

  2. To Change Priority, use the drag-and-drop handle in the PRIORITY column to move policies up or down. Best Practice: Place specific, narrow policies at the top and broad, "catch-all" policies at the bottom.

    For keyboard users, select an item with Space/Enter and move it using the Arrow keys, as indicated in the message at the top of the window.

  3. You can also use the Move Up or Move Down icons to adjust order.

  4. To Activate/Deactivate, click the stacked dots icon in the Action column next to the policy and click Activate/Deactive

  5. To Edit, click on a policy name to modify its scope or filters or click the stacked dots icon in the Action column and click Edit.

  6. Click Save.

Configure a Catch-All Policy

If you want to ensure that all interactions not caught by specific rules are still processed (e.g., for basic redaction or transcription), create a Catch-All policy.

To create a Catch-All policy:

  1. Select your desired Language and Redaction Profile.

  2. Set the Filter Types section to something that captures all interactions like Direction = Inbound and Outbound.

  3. Save the policy and move it to the very bottom of the priority list.

  4. Any interaction that didn't match the specific filters of the policies above it will fall into this policy.

Best Practices for Policy Management

To ensure your analytics run as expected, follow these organizational strategies:

  • Use a Catch-All Policy: Create a policy with no filters and place it at the very bottom of your list (lowest priority). Set it to your most common language and basic redaction. This ensures that any interaction not caught by specific rules is still transcribed.

  • Order by Specificity: Place your most specific policies (e.g., a specific team or high-value skill) at the top of the list. Place broad policies (e.g., "All Inbound") toward the bottom.

  • Consolidate Services: If you want an agent to have both Interaction Analytics (IA) and QM Advanced (QMA) metrics, ensure both toggles are enabled in the same policy. If you split them into two policies, only the higher-priority one will run.

  • Verify Licenses: Toggles for services like Topic AI or Enlighten CSAT only appear if your tenant is licensed for them. If a service is missing from the UI, contact your account representative.

Frequently Asked Questions

Q: Why wasn't an interaction transcribed or analyzed?

A: It either didn't match any policy in your list, or it matched a higher-priority policy that had the analytics toggles turned off. Check your priority order.

Q: Do I still need to check "Can Be Analyzed" on the agent's user profile?

A: Only if your policies are specifically filtered to look for that attribute. If your policy filters by "Team" or "Skill" instead, the individual agent flag is ignored for that policy.

Q: Can I re-run an interaction against a different policy later?

A: No. Policies are applied at the time of processing. Changes to policies only affect new interactions moving forward; there is no reprocessing of historical data.

Q: How does this affect Real-Time interactions (like Copilot)?

A: For real-time transcripts, the matched policy applies its analytics scope and redaction profile to the existing transcript created in real-time. It does not trigger a second, post-call transcription.

Q: An agent has both flags, but I only see IA categories, not QMA?

A: Their segment matched a higher-priority policy that only enables IA. Reorder the policies, or add QMA to that one.

Q: How does this interact with Interaction Hub redaction / Data Policies?

A: Same redaction logic, sourced from the Redaction Profile on the matched policy. "Disable playback until redacted" still works as today (requires Data Policies Advanced license).

Q: A Brazilian customer wants everything Portuguese by default. Can this be handled?

A: There's no "tenant default language" anymore. Create a catch-all policy (no filters) set to Portuguese, plus a higher-priority policy for any English skills.