Search

Minimum required permission: Interaction Analytics (CXone) View

The Search page allows you to search across all your data without being restricted by datasets Set of parsed interaction transcripts that spans a specific period of time and match your specified filter criteria. or categories Data groups that Interaction Analytics (CXone) uses for parsed call transcript data. They make it easier to find trends.. You can search for specific keywords or phrases, segment IDs, contact IDs, team names, or agent names. You can also filter your search by date, language, channel, and other metrics.

The page includes a graph that illustrates trends in interactions over time. The Matching Interactions widget displays interactions that match your search criteria. You can drill down into individual interactions within your search results to learn more.

You can sort the columns in the widget Graphical representation of data that meets specified filter criteria. to customize how they display. To do so, click Columns on the right side of the widget and select the boxes for the metrics you would like to view. Click Columns again to close the window.

Apply Search Filters with AI

Required permission: Interaction Analytics (CXone) Admin

You can use Copilot for Analytics, an AI-powered assistant, to create and refine search filters using natural language prompts. Copilot for Analytics is available on the Search page and helps you quickly find the data you need, even if you are not sure which filters to use.

You must have an Analytics Advanced or Analytics Premium license to use this feature.

Note: Copilot for Analytics is not available in the UAE region.

1. Open Copilot for Analytics:

   • On the Search bar, open the Copilot for Analytics icon .

   • The panel opens on the right side of the page and presents a set of suggested questions you can click to get started.

2. Enter a Prompt or use suggested questions:

   • Type your own prompt

     In the text field at the bottom of the Copilot panel, type a question or describe the interactions you want to find in your own words. For example:

     • "Show me all calls with negative sentiment last week."

     • "Find interactions where the client was frustrated and the issue was unresolved."

     • "List chats in Spanish about payment issues."

   • Use suggested questions

     • Copilot displays a set of example questions in the panel. You can click any of these questions to automatically send it as a prompt.

     • If the suggestions are hidden, click the Copilot icon in the text field at the bottom of the panel to show the set of questions again.

3. Review Suggested Filters:

   Copilot for Analytics interprets your prompt and responds with a list of suggested filters. For each filter, Copilot provides:

   • The filter type and value it selected (for example, Frustration Type: Any Frustration).

   • A brief explanation of why the filter was added based on your prompt.

   The suggested filters are automatically applied as filter chips below the search bar on the Search page (for example, DATE RANGE and LANGUAGE). The Interactions Over Time chart and Matching Interactions grid update immediately to reflect the applied filters.

4. Refine Your Search:

   You can continue to refine results in two ways:

   • Keep prompting Copilot - Type follow-up prompts in the chat panel to narrow or expand your results. For example:

     • "Only show calls from last week."

     • "Exclude resolved interactions."

     • "Add a filter for Spanish language."

   • Edit filters manually — Modify, add, or remove filter chips directly in the filter area, or click + Add Filters to add filters yourself. Copilot preserves your existing filters unless you explicitly ask it to change them.

   You can switch freely between Copilot prompts and manual filter editing at any time.

5. Save Your Search:

   Once you are satisfied with your filters, you can save them as a saved search or dataset filter for future use.

Supported Filters:

Copilot for Analytics can help you set filters for:

• Language

• Client frustration type

• Document rank

• Content type

• Resolution status

• Participant sentiment

• Keywords (include all / include any / exclude)

• Entities (include / exclude, with participant support for Agent, Client, or Either)

• Metrics

• Other standard filters available on the Search page.

All filters applied by Copilot for Analytics respect your role-based permissions, so you only see data you are allowed to access.

Limitations:

• Currently not available in the UAE region.

• You can use natural language prompts only for filters that the AI assistant supports. It does not support filtering by business data fields.

• You must have the appropriate permissions to use Copilot for Analytics and to save searches.

Manually Apply Search Filters

You can filter your search by date range, language, channel, and several other metrics. You can hide the filter options by clicking Filters in the top-right corner.

To remove a filter, click the X next to it, then click the magnifying glass icon . You can also click more options > Reset Widget Filters on an individual widget.

You can enter keywords or phrases into the search bar and click the magnifying glass icon to filter for interactions that only use certain terms. You cannot enter the following characters in the search bar: ' [ ] { } < >. If you enter multiple terms or phrases into the search bar, only interactions that include all the entered terms display. Unless, you want to search for interactions that include multiple terms at once, the Keywords and Phrases filter may be more helpful to use instead. The Keywords and Phrases filter allows you to apply more rules such as exact match and proximity.

1. Click the funnel icon the funnel icon in the top right corner of the Search page. By default, a seven-day date range filter is already applied.

2. To change the date range, click the Date Range drop-down. You can set it to search Within The Last __ Days or set a Custom Date Range by date and time of day. If you set a custom date range, be sure to select one of the 15-minute increments. If you want to select midnight, you can do so by choosing 12 AM on the following day.

3. You can optionally change the Language to filter interactions by. Using this filter is required if you are using more than one language. The selected language defaults to your system's default language.

4. To narrow results by other criteria, click Add Filters and select the type you want to apply. You will need to repeat this step to add more than one type of filter.

5. Click the filter you just added to finish configuring the filter options.

   Learn more about filter options

   The following filters can be applied to analyze just the agent side of interactions, just the client side, or either:

   • Keywords and Phrases

   • Categories

   • Entities

   • Sentiment

   • Company Profile

   For these filters, if you add multiple of the same kind, there is an OR relationship between them. This means that so long as an interaction meets some of the criteria, it's included in the filter results. It doesn't need to meet all the criteria entered to match. For example, if you add two Keywords and Phrases - Agent filters for different words, the agent only needs to have said at least one of the words for the interaction to match.

   If you add multiple different kinds of these filters, there is an AND relationship between them. This means that each item you add to the filter must be present in the interaction to display as a filter result. For example, if you add a Keywords and Phrases - Agent filter for the word awesome and a Keywords and Phrases - Client filter for great, both of these must be included for an interaction to match the filter in this case.

   Filter	Details
   Channels	Filters data by channels used by your organization. Supported channels include: Inbound and outbound Voice Omnichannel and Digital Experience (DX) Chat Omnichannel and Digital Experience (DX) Email SMS. Digital Experience (DX) channels: Apple Business Chat Facebook Messenger In-App Messaging In-App Messaging LINE Messaging Microsoft Teams Slack Telegram Messaging Viber WhatsApp Custom channels using the Bring Your Own Channel feature Chat interactions from Cisco are ingested and analyzed. This includes redacting sensitive data and generating deeper insights into customer and agent text conversations. It helps ensure compliance and provides a better understanding of agent performance.
   Frustration	Filters data based on whether the contact was frustrated Different from negative sentiment. Frustration cues show that a customer is upset rather than just discussing something negative.. Select one of the following Frustration options: Frustrated: Only displays interactions where the client (contact The person interacting with an agent, IVR, or bot in your contact center.) was frustrated. Not Frustrated: Only displays interactions where the client was not frustrated. Note that frustration is different from negative sentiment Overall mood or result of the interaction as determined by analysis of words, phrases, and context of the transcript.. All frustrated interactions should also be negative, but in some cases negative sentiment could mean something else. Frustration cues show that a contact is upset. For example, frustration cues could include words or phrases like , "I'm very angry."
   Readability	Filters data based on how easy it is to read the interaction transcript. Automatic Speech Recognition (ASR) assigns confidence ratings to transcripts based on the number of errors they have. The fewer the transcript errors, the easier it is to read. So, this is helpful if you have to read many transcripts to solve an issue and want to focus on transcripts that are easier to read. By default, all transcripts are included. Select one of the following Readability options for an ASR confidence rating: High: Only displays interactions with transcripts that are easier to read. Low: Only displays interactions with transcripts that are more difficult to read. Important: Filtering by High confidence will likely still return transcripts with errors, but they will be easier to read than lower quality transcripts.
   Resolution	Filters data based on whether the cause for the interaction was resolved during the interaction. Select one of the following Resolution options: Resolved: Only displays data for resolved interactions. Unresolved: Only displays data for unresolved interactions.
   Keywords and Phrases	Includes or excludes interactions from the data based on keywords or phrases Combination of words that have special significance when used together in a specified order, such as "want to cancel". they contain. You can enter a keyword or phrase to return exact matches or variations. For phrases, you can also specify how close the words in the phrase must be to each other for an interaction to be considered a phrase match. You can choose to view data from only the agent side, only the contact side, or both sides of interactions. If you only filter for one side of interactions, you can still see both sides when you drill down to the transcript view. Select one of the following options: Keywords and Phrases - Agent: Only displays agent data. Keywords and Phrases - Client: Only displays client (contact The person interacting with an agent, IVR, or bot in your contact center.) data. Keywords and Phrases: Displays both agent and client data. Enter keywords or phrases you want to filter for. Select one of the following Keywords and Phrases options from the first drop-down: Have: Displays data for interactions that mention the word or phrase you enter. Have at least one: Displays data for interactions that mention at least one of the terms or phrases you enter. This is only helpful if you use more than one keyword or phrase filter. Not have: Only displays data for interactions that don't mention the term you enter. Select one of the following options from the Matching drop-down: Exact: Only displays interactions that include the term or phrase exactly as you enter it. Approximate: Displays interactions that include variations of the term or phrase you enter. For example, if you enter activate, it could display interactions that include activate, activates, activating, activated, and so on. If you entered a phrase, use + and - or enter an integer to configure Proximity. This includes matches where the words in the phrase aren't consecutive. Phrase matches can have up to eight words between them. The default is 0. For example, the phrase speak manager with a proximity of 2 would capture the phrases speak to a manager, speak with your manager, and so on. Tip: Including commas and apostrophes in your search can affect your results. To avoid this issue, set a Proximity value when using the Keywords and Phrases filter. Click Add when you're finished. Repeat these steps to add additional filters.
   Categories	Filters data by existing category rules. You can choose to view data from only the agent side, only the contact side, or both sides of interactions. If you only filter for one side of interactions, you can still see both sides when you drill down to the transcript view. Select one of the following options: Categories - Agent: Only displays agent data. Categories - Client: Only displays client (contact The person interacting with an agent, IVR, or bot in your contact center.) data. Categories: Displays both agent and client data. Drill down and select the categories you want to filter for. Selected categories become blue.
   Entities	Includes or excludes interactions from the data based on entities Keyword or phrase defined in your company profile in Interaction Analytics (CXone). Related to an entity type. Can include variants. they contain. Some entities are preconfigured in Interaction Analytics (CXone). If you want to create custom entities, you must configure them in the company profile. If you want to filter based on keywords and phrases, use that filter instead. You can choose to view data from only the agent side, only the contact side, or both sides of interactions. If you only filter for one side of interactions, you can still see both sides when you drill down to the transcript view. Select one of the following options: Agent: Only displays agent data. Client: Only displays client (contact The person interacting with an agent, IVR, or bot in your contact center.) data. Either: Displays both agent and client data. Select whether you want to Include Entities or Exclude Entities, then choose the entities you want to include or exclude.
   Sentiment	Filters data by sentiment Overall mood or result of the interaction as determined by analysis of words, phrases, and context of the transcript.. You can choose to view data from only the agent side, only the contact The person interacting with an agent, IVR, or bot in your contact center. side, or both sides of interactions. If you only filter for one side of interactions, you can still see both sides when you drill down to the transcript view. Select one of the following options: Agent: Only displays agent data. Client: Only displays client (contact The person interacting with an agent, IVR, or bot in your contact center.) data. Either: Displays both agent and client data. Select one of the following sentiment types: Overall Sentiment: The sentiment for the whole interaction. Beginning Sentiment: This is determined by the first 400 words, or first 30%, of the interaction, whichever occurs first. It can help you identify common reasons clients call about. Mixed sentiment is not available for beginning sentiment. End Sentiment: End sentiment is determined by the last 30% of an interaction. Mixed sentiment is not available for end sentiment. Once, you've selected a sentiment type, choose one or more of the following sentiments: Positive: The number of positive cues is greater than the number of negative or neutral cues in an interaction. Negative: The number of negative cues is greater than the number of positive or neutral sentiment cues in an interaction. Note that negative sentiment is different from frustration Different from negative sentiment. Frustration cues show that a customer is upset rather than just discussing something negative.. Some phrases can be both negative and frustrated. However, a client (contact The person interacting with an agent, IVR, or bot in your contact center.) may be discussing a negative issue they want addressed without necessarily being frustrated. Neutral: An absence of negative or positive cues. May also occur if you have phrases set as neutral in your company profile. Mixed: The number of positive and negative cues in an interaction are essentially equal. Doesn't appear as an option for Beginning Sentiment or End Sentiment. If you don't select a sentiment, the default is to include interactions of all sentiments.
   Company Profile	Filters for entities Keyword or phrase defined in your company profile in Interaction Analytics (CXone). Related to an entity type. Can include variants. configured in your company profile Page in Interaction Analytics (CXone) where you can create custom entities.. You can filter by the entity type, like Products, or drill down to include entities within the entity type, like Sweaters. The options will vary depending on the entities and entity types you have added to your company profile. You can choose to view data from only the agent side, only the contact side, or both sides of interactions. If you only filter for one side of interactions, you can still see both sides when you drill down to the transcript view. Select one of the following options: Agent: Only displays agent data. Client: Only displays client (contact The person interacting with an agent, IVR, or bot in your contact center.) data. Either: Displays both agent and client data. To include all configured entities under a single entity type, such as Company, select the checkbox beside the entity type name. Clear the checkbox to remove it. To select only certain entities under an entity type, click the filter you just selected again and select the checkbox next to each entity you want to include. Clear the checkbox to remove it.
   Metrics	Check the box next to the specific metric type you would like to filter for. After you check the box for the metric you want, click the filter you just added to select a value for the specified data field. Click the check mark to the right when you are finished. For example, if you select Agent Name as a metric, you would need to enter or select the name of the agent you would like to filter for. See the Metrics entry in Key Terms and Metrics for more information about each available metric type.
   Business Data	This filter is only visible if your system has Business Data (custom metadata) fields configured. It allows you to filter interactions based on additional attributes imported from external systems, such as CRM, IVR, or other business applications. These fields can include information like customer region, product code, escalation status, or any other custom data your organization defines. Check the box next to the business data field you want to include in your search. Then, select or enter one or more values for that field. Only interactions that match your selected business data criteria will be shown in your results. Business Data fields support multiple data types, including text, numeric, and boolean values, and can be combined with other filters for more targeted analysis. The available fields and values depend on your organization’s configuration. You can define up to 50 business data fields per tenant (35 text, 10 numeric, 5 boolean). You can combine multiple business data filters for more targeted results. For example, filter for interactions where Customer Region is "EMEA" and Is Escalated is true.   You can also use the Copilot for Analytics panel to filter by business data fields using natural language prompts. For example, type: “Show me interactions where Product Code is 12345 and Is Escalated is false.”
   Customer Satisfaction	This filter is only visible if your system enables NiCE CXone AI features. It filters for specific NiCE CXone metrics and the scores for those metrics. These metrics and scores measure behaviors that demonstrate efforts to achieve customer satisfaction. Check the box next to the customer satisfaction behaviors you want to include in your search. Then select one or more scoring level for the metrics you want to see. The scoring levels include: Moderately Negative, Neutral, Moderately Positive, Strongly Positive.
   Sales Effectiveness	This filter is only visible if your system enables NiCE CXone AI features. It filters for specific NiCE CXone metrics and the scores for those metrics. These metrics and scores measure behaviors that demonstrate efforts to make sales effectively. Check the box next to the sales effectiveness behaviors you want to include in your search. Then select one or more scoring level for the metrics you want to see. The scoring levels include: Strongly Negative, Moderately Negative, Neutral, Moderately Positive, Strongly Positive.
   Intents	​Filters by the intent The meaning or purpose behind what a contact says/types; what the contact wants to communicate or accomplish. of the interaction as assigned by NiCE CXone. This allows you to view interactions by purpose based on what the contact wants to communicate or accomplish. Only one intent is assigned to an interaction, but you can filter by multiple intents at once. Intents are sorted into three different levels: category, topic, and intent. If you select the box next to a category, it also filters by the topics and intents included in that category. Intents are organized by topic, and topics are organized by category. If you only want to filter by certain intents, click the drop-downs of the relevant category and topic to drill down to the desired intent. It may be helpful to create categories by intent. This is supported for voice and digital channels. Supported digital channels include all conversational digital interactions but not social media posts. They also include legacy ACD chat and SMS channels. Legacy ACD email and NiCE CXone email are not supported. To use this, you must have one of these licenses or license combinations:  Analytics Advanced Analytics Premium Analytics Essential and Automated Summary (AutoSummary) If you have questions, ask your account representative.
   Outcomes	Filters by the outcome Result assigned by the agent or system at the end of a voice (disposition) or digital (status) interaction. of an interaction. For example, outcomes could include payment completion, a transfer, or an escalation. Interactions can have up to 12 outcomes. Select the box next to the outcomes you want to filter for. It may be helpful to create categories by outcome. This is supported for voice and digital channels. Supported digital channels include all conversational digital interactions but not social media posts. They also include legacy ACD chat and SMS channels. Legacy ACD email and NiCE CXone email are not supported. To use this, you must have one of these licenses or license combinations:  Analytics Advanced Analytics Premium Analytics Essential and Automated Summary (AutoSummary) If you have questions, ask your account representative.
   Actions	Filters by actions that occurred during an interaction. For example, actions could include processing a payment, providing account information, or pulling up an account. Interactions can have up to eight actions. Select the box next to the outcomes you want to filter for. It may be helpful to create categories by action. This is supported for voice and digital channels. Supported digital channels include all conversational digital interactions but not social media posts. They also include legacy ACD chat and SMS channels. Legacy ACD email and NiCE CXone email are not supported. To use this, you must have one of these licenses or license combinations:  Analytics Advanced Analytics Premium Analytics Essential and Automated Summary (AutoSummary) If you have questions, ask your account representative.

6. Once you have finished configuring any filters, click the magnifying glass icon .

Save a Search

You can select saved searches from the drop-down and quickly view results without reapplying filters. Saved searches can be public or private. Public searches are visible to all users. Private searches are only visible to the user that created them or users with the Admin permission.

To manage saved searches, click more options > Saved Searches in the top-right corner. Here you can view details, copy, and delete saved searches. You need the Admin permission to delete searches created by other users.

1. Click the app selector and select Analytics.
2. Go to AutoDiscovery.

3. To save a search, configure the search settings and filters for the data you would like to view. Then, enter your search.

4. When you are finished, click the save icon in the top right corner.

5. In the New Search field, enter a name for your search. It may be helpful to include numbers or special characters in the name, but not all characters are supported. For example, you could use the names Kansas Team > 30 seconds; Team Emerald City or Team Munchkinland 'Escalations' [with transfer].

   You will get an error message if you use the following characters to name a saved search: a comma, a period, #, ? , $, %, or ( ).

   Supported characters:

   • Alphanumeric characters (A-Z, uppercase and lowercase, and 0–9)
   • Underscores ( _ )
   • Spaces
   • Hyphens ( - )
   • Colons ( : )
   • Semicolons ( ; )
   • Double quotes ( " " )
   • Single quotes ( ' ' )
   • Pipe symbols ( | )
   • Less than and greater than symbols ( < > )
   • Square brackets ( [ ] )
   • Curly brackets ( { } )

6. To select a different color for your search, click the colored box in the New Search field and select another color.

7. Enter a description of your search settings and filters.

8. Enable Public, to make the search visible to other users. Saved searches are private by default.

9. Click Save As. Now your search is available in the drop-down and you can reuse the same search criteria later.

Create a Topic AI Collection

Required permission: Interaction Analytics (CXone) Admin

You can push search interaction results from IA directly to Analytics Hub as a Topic AI collection. You can name your collection and submit up to 40,000 sample segment IDs to the Topic AI API. This makes creating and managing collections faster and easier. It also eliminates manual data collection to make workflows more efficient.

Collections are created in Interaction Analytics (CXone), but you must go to Topic AI to view the collection after it's created. It takes approximately 20–30 minutes before it appears in the Topic AI > Select Topic AI Collection dropdown during enrichment.

You must have an Analytics Advanced or Analytics Premium license to use this feature.

1. On the top right corner of the Search page, click > Save to Collection.

2. Enter a New Collection Name.

   1. Allowed characters: Letters, numbers, spaces, #, and @

   2. Following characters are not allowed:

      • Commas (,)

      • Semi-colon (;)

      • Colon (:)

      • Underscore ( _ )

      • Pipe " | "

      • Double Quote ""

      • Single Quote " ' "

      • Left and Right square bracket " [ ] "

      • Left and Right curly brackets " { } "

      • Less than and Greater than " < > "

      • ?, $, %, or ( ),

      If the value contains disallowed characters, a validation message appears, and you must correct the value before saving.

3. Click Confirm.