About Analytics tasks

On this page:

General sentiment definitions

The table below defines the possible sentiments that RingCentral Analytics can assign to the agent and the customer for a contact.

Sentiment Definition
Positive The contact shows clear signs of satisfaction, resolution, or being a supportive and helpful conversation. These signs might include appreciative language, smooth issue resolution, or a generally upbeat interaction.
Neutral The interaction is professional and factual throughout, with little emotional expression, neither strongly positive nor negative. Contacts with neutral sentiment are typically routine or transactional conversations.
Negative The conversation reflects frustration, complaints, unresolved issues, or tension. It could include repeated expressions of dissatisfaction, escalations, or language that shows irritation or disappointment.

Example custom sentiment definitions for Sentiment Analysis tasks

If your organization has an Analytics Enterprise Plus license, you can define what positive, negative, or neutral sentiment looks like for your organization as part of creating a Sentiment Analysis task. Below are some example definitions.

Sentiment Sample custom definition
Positive The patient expresses clear satisfaction with the contact center experience. This includes feeling helped or reassured; thanking the agent; describing the interaction as helpful, easy, or professional; or expressing relief, gratitude, or appreciation. Example phrases: “Thank you so much for your help.” “You were very kind and explained everything clearly.” “This was really easy. I appreciate it.”
Neutral The patient shows no strong emotion, either positive or negative. This includes simply acknowledging information, responding factually or briefly without praise or complaints, or ending the call politely but indifferently. Example phrases: “Okay, got it.” “Thanks.” (without additional emotion or context) “That answers my question.”
Negative The patient expresses frustration, confusion, or dissatisfaction with the interaction, even if the medical issue is unrelated. This includes feeling ignored, misunderstood, or rushed; expressing irritation, sarcasm, or disappointment; or indicating unresolved issues or the need to call again. Example phrases: “This was a waste of time.” “You’re not listening to me.” “I’ve called three times and still no help.”

Example custom topics and descriptions for Trending Topics tasks

If your organization has an Analytics Enterprise Plus license, you can define your own contact topics as part of creating a Trending Topics task. Below are some examples of topics and descriptions.

Topic name Topic description
Order cancellation: Delayed delivery The customer is requesting to cancel their order because the delivery has been delayed beyond the expected timeframe.
Schedule appointment The patient is booking a new appointment for a clinical visit, procedure, or consultation.
Missed appointment follow-up The contact was initiated after a patient no-showed their appointment. The purpose of the contact is to reschedule or discuss next steps.
Prescription refill request The patient is requesting a refill for an existing prescription.
Open new account The customer is requesting to open a new checking, savings, or other account.

Upload intervals for Desktop Analytics data

Desktop Analytics data uploads at different intervals based on the type of recording system your organization uses.

Desktop Analytics data is uploaded and processed each time the agent turns on or turns off their computer, or at midnight local time (based on the computer’s system time zone) if the computer is running. If the agent puts their computer to sleep instead of shutting it down, Desktop Analytics data does not upload into Calabrio for RingCX.

Condition definitions

For several Analytics tasks, you can set the conditions that must be met for a contact to be processed by the task. The conditions available vary based on the contact type. The table below defines the conditions.

Contact detail Definition Available for these contact types
Call Direction Whether the contact was inbound or outbound. Chat, Email, Call

Calling In Number

 The phone number of the party who initiated the call. For inbound calls, this is the customer’s phone number. For outbound calls, this is the agent’s or organization’s phone number. Call
Calling out number The phone number of the party who received the call. For inbound calls, this is the agent’s or organization’s phone number. For outbound calls, this is the customer’s phone number. Call
Contact Duration (seconds) The number of seconds from when the contact was picked up to when the contact concluded. Chat, Email, Call
Channel The system through which the contact was received. Chat, Email
Email To The email address of the recipient. For inbound emails, this is the agent or organization’s email address. For outbound emails, this is the customer’s email address. Email
Email From The email address of the sender. For inbound emails, this is the customer’s email address. For outbound emails, this is the agent’s or organization’s email address. Email
Email Subject The subject line of the email. This text is the same for both inbound and outbound emails. Email

Changes to Analytics licenses

If your organization downgrades from an Analytics Essentials, Analytics Enterprise, or Analytics Enterprise Plus license to a Desktop Analytics license, Calabrio for RingCX automatically disables all existing non-desktop tasks. This disabling does not affect any tasks that are in progress—for example, ad-hoc tasks that have already started or ongoing tasks that are scheduled or started.

Email and chat text

You can set up an ongoing Text Analytics task to analyze emails for predefined phrase hits. If no ongoing task is configured, then all emails are still indexed and saved as email contacts. You can process those emails with ad hoc tasks.

You must set up a Phrase Hits - Digital Contacts task or a Text Analytics task to analyze emails or chats for predefined phrase hits. If no task is configured, here is what happens when text is imported into Calabrio for RingCX:

  • The text is not analyzed in Analytics.
  • The contact is still available on the Interactions page.
  • Sentiment is not analyzed on the Interactions page.

NOTE   If Calabrio for RingCX detects a generic email address that is not associated with an agent during an email import to the Mail server, Calabrio for RingCX creates a new user and associates the user with that email address. This user then appears on the user list and adds a user to the user count.

Languages and teams

For a complete list of supported languages in Calabrio for RingCX, see Localization and supported languages.

IMPORTANT   Languages require server configuration. Before using a language in a task, make sure your system administrator has configured the language.

When you run a speech task, results are based exclusively on the selected language. There is an implied relationship between teams and languages. For example, if you choose French from the Language drop-down list when configuring a task, the task processes all calls in French for the selected teams. Any secondary language spoken during those calls produces bad results.

BEST PRACTICE   

If your contact center supports multiple languages, each team should be associated with a specific language.

It is not unusual for one or more agents in a team to be bilingual. In that case, bilingual agents switch between languages during a call or from one call to another. If agents on a team are known to use a secondary language during a call, create a metadata tag using the Metadata Manager page (Application Management > QM > QM Configuration > Metadata Manager) and assign the possible language values to a metadata tag. When an agent uses a secondary language during a call, the agent can assign that metadata tag to identify the language used. You can then run a task to search for all calls associated with the team that are tagged with that specific language metadata tag.

EXAMPLE   

Agent A is on the Sales team and speaks French as a primary language and English as a secondary language. When Agent A switches to English for a call, she tags the call with a metadata tag called VoiceLanguage and a value of English.

To analyze Agent A’s English language calls, you create a Speech to Text (Transcription) task configured with the following settings:

  • Language: English
  • Team: Sales

If you select French from the Language drop-down list, the task will not yield any results.

Related topics