Sprinklr Service unifies customer experiences across voice, digital, and social channels. This release delivers advanced capabilities, enhancing efficiency and enabling businesses to craft consistent interactions across all touchpoints. Here are the key features included in Sprinklr Service's latest release.

Summary

Feature Updates

The following feature updates are being introduced in Sprinklr Service:

After Call Work (ACW)

The following features are being introduced in Sprinklr Service’s After Call Work (ACW) module:

Simulate ACW

You can now simulate After Call Work (ACW) directly from the ACW Manager using the new Execute option. This
enhancement allows administrators and configuration teams to test ACW flows without making real calls or using agent capacity.

Previously, validating an ACW required placing test calls, which was time‑consuming and resource‑intensive. With in‑app simulation, you can quickly verify ACW behavior during setup, identify configuration issues early, and reduce dependency on live testing.

Key Capabilities

• Execute ACW from ACW Manager - A new Execute option is available for each active ACW. Draft ACWs cannot be executed.

• In‑place simulation experiences - The ACW opens in a modal on the same tab, similar to an expanded ACW view—no new tab or pending ACW section is required.
• Simulation inputs - Provide a Case Number and select a Conversation ID (with inbound/outbound indicators and channel icons where applicable).
• No impact on operations - Simulation does not consume agent capacity and does not require a live call.
• Controlled execution - No ACW timer, redial, retry, extensions, or auto wrap are applied during simulation. You can exit the simulation at any time.
• Error visibility - If a simulation fails, the error code and description are displayed, and the failing node is highlighted in the ACW flow editor to help with debugging.

Permission - Access to simulation is governed by the new Execute ACW permission.

Availability - Available for all active ACWs in ACW Manager.

For further details, see ACW Manager - UI capabilities.

ACW Redial in Reporting

Sprinklr has enhanced how After Call Work (ACW) Redial interactions are attributed to work queues in reports. ACW redials are now consistently reported against the last associated queue from the original call, instead of being defaulted to Dummy Workqueue when a valid queue exists.

This update ensures more accurate queue‑level insights and agent performance reporting.

​

​

What's Changed

Previously, ACW redial interactions often appeared under Dummy Workqueue, even when the original call was routed through a queue.

With this update:

• ACW redial reporting is based on the original call that entered ACW
• The last associated queue for the agent who triggers the redial is applied
• If the original call had no queue association, reporting continues to show Dummy Workqueue
• ACW redial does not create or modify queue associations

Only the queue name or Dummy Workqueue is displayed for ACW redial interactions.

Example

A call enters Queue 1 and is handled by Agent 1.
The call is then warm‑transferred to Queue 2 and handled by Agent 2.

• If Agent 1 initiates an ACW redial, reports show Queue 1
• If Agent 2 initiates an ACW redial, reports show Queue 2

Each agent’s ACW redial reflects their own last associated queue.

​

Reports Impacted

This improvement applies to ACW redial data across the following reports:

• Agent Voice Performance Report
• Voice Report (Customer)
• Voice Queue Performance Report
• Omni‑Channel Queue Performance Report
• Agent Detailed Call Transfer Report
• Agent Occupancy
• Omni‑Channel Case Summary Report

For further details, see ACW Reporting.

Agent Copilot

Support for BYOM Streaming

Agent Copilot now supports streaming for Bring Your Own Model (BYOM) integrations. When streaming is enabled, answers appear in real time as they are generated during on-demand requests, instead of waiting for the full response to complete. Streaming reduces perceived wait time and helps you access information faster during conversations. You can start responding with confidence while Copilot finishes generating the answer.

For further details, refer to Streaming Support in Agent Copilot.

Clone Existing Agent Copilots

You can now clone an existing Agent Copilot directly from the vertical ellipses (⋮) menu in the Agent Copilot Record Manager. The clone includes supported configuration settings such as data sources, tasks and tools, LLM settings, widget configurations, and conditions. Cloned Copilots are created in an Inactive state, giving administrators an opportunity to review and adjust the configuration before activation. This enhancement simplifies the creation of similar Copilots and reduces configuration effort. For further details, refer to Available Actions in Agent Copilot.

Enhanced Proactive Assistance for Agents

Agent Copilot now provides enhanced proactive assistance during live customer conversations by automatically identifying key interaction opportunities and surfacing timely recommendations to agents. The following features are being introduced:

​

• Promise Management helps agents track customer commitments by automatically detecting promises made during conversations, such as callbacks or follow-ups. When a promise is identified, Copilot displays an editable Promise Management card that agents can review, update, approve, and save, helping ensure commitments are accurately captured and fulfilled.

• Proactive Upsell Recommendations identify potential upsell opportunities based on conversation context and customer sentiment. When an opportunity is detected, Copilot surfaces an Upsell card with relevant recommendations and a contextual message draft that agents can review before sending. Agents can choose to pitch the suggested offer or dismiss the recommendation as needed.

To improve visibility of proactive guidance, Agent Copilot now also generates platform notifications whenever a configured proactive suggestion or task is triggered during an active customer conversation.

For further details, refer to Agent Copilot Capabilities.

Add External Websites as Knowledge Source

Agent Copilot now supports indexing content from external websites as a knowledge source. A new Websites knowledge source option in Data Management allows organizations to index supported external websites alongside existing sources, such as Knowledge Base articles, canned responses, and platform cases.

By expanding the range of available knowledge sources, this enhancement helps Copilot generate more comprehensive and contextually relevant responses.

For further details, refer to Configure Websites as Knowledge Source in Agent Copilot.

Callbacks

The following features are being introduced in Sprinklr Service’s Callback module:

Default Callback Number Selection in Callback Manager

This release improves the callback experience by ensuring that the customer-provided callback number is selected by default when agents manually dial callback tasks from the Callback Manager.

Customers can provide a preferred callback number through DTMF input in the IVR, and this number is stored at the task level when the callback task is created. Previously, although the Callback Dialer correctly used the callback number stored in the task, the Manual Task Dialer displayed the customer profile phone number by default. This could require agents to manually select the correct number and increase the risk of calling the wrong contact.

With this enhancement, when the dial icon is accessed from the Callback Manager, the system prioritises the callback number stored in the callback task and displays it as the default selected number in the Manual Task Dialer. The profile phone number remains unchanged and independent, ensuring that callback-specific information does not overwrite customer profile data.

This change helps agents initiate callbacks more accurately, reduces manual effort during callback handling, and delivers a more consistent callback experience. The enhancement applies only to the Callback Manager and does not affect dial icons available in other areas of the platform, such as Care Console or Reporting DB.

Agents handling callback tasks through the Callback Manager can use this enhancement whenever initiating a manual callback. The customer-provided callback number will automatically be preselected, allowing agents to quickly verify and dial the intended callback number. For further details, see Callback Actions.​

Call Control

The following features are being introduced in Sprinklr Service’s Call Control module:

Introduction of New Metrics in Voice Report for Twilio Provider Queue Time

Voice Reports now provide visibility into Twilio provider queue time for both agent and customer call legs. Twilio Queue Time, returned by Twilio during call creation or conference participant creation, is now captured, stored, and available for reporting.

This enhancement helps administrators and supervisors understand how long calls wait in the Twilio provider queue before being connected, providing additional insight into provider-level call delays.

The following metrics are now available in Voice Reports:

• Agent Leg - Twilio Queue Time (Provider) – Measures the time an agent call leg spends waiting in the Twilio provider queue. This metric is applicable when an agent initiates a call to a customer.
• Customer Leg - Twilio Queue Time (Provider) – Measures the time a customer call leg spends waiting in the Twilio provider queue. This metric is applicable when a customer calls a toll-free number or when a customer call leg is created through outbound dialing workflows.

Note: Twilio Queue Time represents the time spent in the Twilio provider queue and is different from Sprinklr queue metrics, such as Queue Time, which track queue activity within the Sprinklr platform.

For further details, see Voice Report.

Supervisor Ability to Take Over a Live Call

Supervisors can now take ownership of an active customer call while monitoring agent interactions. A new Take Over Call option has been added to supervisor call controls, enabling supervisors to seamlessly transition from monitoring to handling the conversation directly.

This feature is useful for supervisors who monitor live calls.

After a supervisor starts listening to an active call, they can choose from the following intervention options:

• Whisper
• Barge-In
• Take Over Call

When a supervisor selects Take Over Call:

• The agent is removed from the call.
• The supervisor becomes the primary participant.
• The customer remains connected without interruption.

This release also strengthens supervisor call controls during monitored calls.

 Agents can no longer:

• Remove supervisors from a monitored call.
• Mute supervisors.
• Place supervisors on hold.

Permission Controlled - The Take Over Call capability is controlled through the Voice Take Over Call permission. The option is displayed only for supervisors who have this permission assigned.

For further details, see Call Controls Functionalities for Supervisor Persona.

Keyboard Shortcuts for Call Controls Functions

You can now open the following call control forms directly by using keyboard shortcuts while on an active call:

• Open Transfer Call window - Option + Shift + T (Mac), Alt + Shift + T (Windows)

• Open Conference Call window - Option + C (Mac), Alt + C (Windows)

• Open Record Call window - Option + R (Mac), Alt + R (Windows)

Additional Enhancements

• Keyboard shortcuts are available only during an active call.
• When a form opens through a keyboard shortcut, focus automatically moves to the first interactive element, improving keyboard navigation and accessibility.
• You can close the opened form by pressing Esc.

For further details, see Keyboard Shortcuts for Basic Call Controls Functionalities.

Introduction of New Metrics in Agent Voice Performance & Omni Channel Queue Performance Reports

The Agent Voice Performance and Omni Channel Queue Performance Reports now include additional hold and consult related metrics to provide deeper visibility into how customers are placed on hold during a
call and offers detailed insights into consult interactions between agents.

• Total Consult Handle Time (Agent) - Total time the consulting agent spends connected to the original agent during a consult, including talk time and any hold time during the consult interaction.
• Total Consult Talk Time (Agent) - Total time the consulting agent spends actively talking to the original agent during a consult interaction.
• Total Consult Hold Time (Agent) - Total time the consulting agent is placed on hold by the original agent during a consult interaction.
• Total Consult Wrapup Time (Agent) - Total time the consulting agent spends in after-call work after the consult interaction with the original agent ends.

For further details, see Voice Agent Performance Report and Omnichannel Queue Report.

Care Console

The following features are being introduced in Sprinklr Service’s Care Console module:

Control Drag‑and‑Drop Attachments with Permissions

Drag‑and‑drop and copy/paste uploads in the Care Console reply box are now governed by the Desktop Upload permission within the Media Uploader. Agents without this permission cannot upload files directly from their desktop, while DAM attachments remain fully supported. This is applicable for non-Email channels.

For the email channel, attachment handling is managed through Record Page editor settings:

• Toggle Enable Drag & Drop Attachments for Email on/off

• Toggle Enable Media Uploader for Email on/off

These enhancements enable brands to ensure secure and compliant workflows by requiring agents to send attachments exclusively from the Digital Asset Management (DAM) system, preventing uploads directly from their devices.

For further details, see Drag-and-Drop and Copy-Paste File Uploads.

Ability to Remove Default Case Stream Filters

From the Persona App settings, administrators can now remove the default Case Stream fixed filters (Duration, Status, Priority, and Case Type). Instead of relying on fixed filters, administrators can configure which filters are displayed directly through Persona App settings.

This enhancement makes the filter panel customizable, showing only the filters that are relevant to each team’s workflow. Agents see only the filters that matter to their work, resulting in faster case navigation and a cleaner interface.

For further details, see the Case Stream Filter Fields setting in the Case Stream table.

Support for Showing Tag Names in Case Tags

Case tags in the Care Console now support displaying tag names in addition to the existing color or image identifiers. This enhancement improves readability and allows agents to quickly recognize tag context without relying solely on color codes. For example, when reviewing a case card in the Case Stream, tags that once appeared only as colored dots can now display meaningful labels such as “High Value Client.”

For further details, see the Show Tag Name in Case tags setting in the Case Stream table.

Enhanced Selected Case Indicator

The selected case indicator in Case Stream has been updated from a faint grey box to a high‑contrast marker, making the active case easier to spot in busy workstreams.

Community

The following features are being introduced in Sprinklr Service’s Community module:

Anonymous Community Visitor Metrics Available Regardless of Cookie Consent

For Communities with Sprinklr Analytics enabled, the following anonymous metrics are now tracked regardless of a visitor’s cookie consent preference:

• Page Visits

• Sessions

• Session Duration

• Unique Visitors.

These metrics are also available in Reporting and Analytics as anonymous users. Because these metrics are anonymous in nature, this enhancement provides more complete Community insights while respecting user privacy. For more details, refer to Cookies Used by Sprinklr Community.

Conversational Analytics

The following features are being introduced in Sprinklr Service’s Conversational Analytics module:

Provide Feedback on Detected Topics 

You can now provide feedback for detected Topics at the interaction level, enabling more precise and actionable feedback for individual conversations. This enhancement allows users to provide feedback directly within the application, helping improve topic detection accuracy and relevance. 

For each detected Topic, you can: 

• Like the topic when it is correctly identified. 

• Dislike the topic when it is incorrect or not relevant. 

• Report missing topics that should have been detected. 

Feedback is associated with individual interactions rather than the overall case. This allows users to accurately review and evaluate topics when multiple interactions within the same case contain different conversation contexts and detected topics. For further details, refer to Provide Feeback on Detected Topics.

Conversational AI

Zero-shot debug logs have been enhanced with additional execution details, making it easier to troubleshoot issues and analyse performance.

• Execution End Time: Displays when the zero-shot request completed processing.
• Execution Time: Shows the total time taken to process the request.
• Model: Identifies the LLM used to execute the zero-shot request.

These enhancements provide greater visibility into execution behaviour, helping users debug workflows more efficiently.

Guided Workflows

The following features are being introduced in Sprinklr Service’s Guided Workflows module:

Export Support for Matrix Table Layouts

This release introduces export functionality for Matrix Table layouts, enabling users to download and share data directly from matrix-based views.

Previously, Matrix Table layouts did not provide an export option, limiting users' ability to extract data for offline analysis, reporting, and collaboration. Users had to rely on alternative reporting methods to access information displayed in matrix tables.

With this enhancement, users can export data directly from Matrix Table layouts, including views configured with filters. The export capability is also supported in external contexts, providing a more consistent reporting experience across different deployment scenarios.

Analysts, supervisors, and business users can leverage this capability to simplify data sharing, perform offline analysis, and generate reports from Matrix Table views more efficiently.

On Change Actions for Checkbox Components in Guided Workflow

This release introduces On Change Actions for Checkbox components in Guided Workflow, enabling workflows to respond automatically when a checkbox is selected or deselected.

Previously, Checkbox components did not support action-based navigation, limiting builders to components such as Picklists and Dynamic Tables for triggering workflow actions. This reduced the flexibility of workflows that relied on checkbox interactions.

With this enhancement, builders can configure an On Change Action for Checkbox components and associate Next or Back Screen actions that execute whenever the checkbox state changes. Actions are triggered for both selection and deselection events, including scenarios where a checkbox is pre-selected by default and later modified by the user. Additionally, the checkbox API name is captured in the BUTTON_ACTION process variable when navigation is initiated through a checkbox-triggered action.

Workflow designers and administrators can use this capability to create more dynamic and responsive workflow experiences while reducing the need for additional navigation controls. The existing behaviour of Checkbox Group components remains unchanged.

Adjustable Popover Height for Hierarchical Pick Lists

This release introduces support for configuring the popover height of Hierarchical Pick List components, providing greater flexibility when displaying hierarchical data within Guided Workflows.

Previously, Hierarchical Pick Lists used a fixed popover height, which could make it difficult to view and navigate large or deeply nested hierarchies without excessive scrolling. Administrators had limited control over how much content was visible within the selection window.

With this enhancement, administrators can configure the popover height of Hierarchical Pick Lists to better accommodate their hierarchy structure and user experience requirements. This allows more options to be displayed within the pick list, reducing scrolling and improving navigation through hierarchical data.

Workflow designers and administrators can use this capability to optimise the display of hierarchical selections, making it easier for users to browse, search, and select values within complex hierarchies.

Date & Time Input Enhancements in Guided Workflow

This release enhances the Date & Time Input component in Guided Workflow by improving time selection flexibility, mobile usability, and localisation support.

Previously, users could select time values only in fixed 15-minute intervals, and mobile browsers could exhibit inconsistent behaviour when interacting with the time picker. In addition, localisation requirements could result in less intuitive button labels across translated experiences.

With this enhancement, administrators can configure custom Time Intervals to control the time slots available for selection. The update also improves browser compatibility by ensuring that time options are immediately scrollable when the picker is opened on supported mobile browsers, including Samsung Internet, Chrome Mobile, and Safari Mobile. Additionally, updates to button labelling provide greater flexibility for localisation and translation scenarios.

Workflow designers and administrators can use these enhancements to create more flexible scheduling experiences, improve mobile workflow completion, and deliver a more consistent experience across browsers and languages.

Support Account Association During Case Creation in Guided Workflow

This release enhances the Create Record node in Guided Workflow by enabling support for the Account field when creating Case records.

Previously, while users could retrieve the Account associated with an existing Case, they could not populate the Account field during case creation. As a result, newly created cases were not automatically associated with the intended Account, creating inconsistencies between case creation and retrieval experiences.

With this enhancement, users can now pass an Account ID as part of case creation. The system persists the Account value and automatically associates the newly created Case with the specified Account, ensuring consistency between read and write operations for the Case entity.

Workflow designers and administrators can use this capability to create source-agnostic cases with accurate account associations, improving data consistency and supporting account-based customer service workflows.

File Preview and Download Support in Attach File Component

This release enhances the Attach File component in Guided Workflow by adding support for file preview and download, allowing agents to verify uploaded documents before proceeding with workflow actions.

Previously, files uploaded through the Attach File component were displayed only as attachments, with no option to preview or download them directly from the workflow. This made it difficult for agents to confirm that the correct document had been uploaded, particularly when the file was intended for customer-facing communications.

With this enhancement, agents can open and preview supported file types, such as PDFs, directly from the Attach File component. Uploaded files can also be downloaded for verification when needed. The preview experience leverages existing platform capabilities to provide a consistent and seamless document review experience while preserving the current file upload behaviour.

Agents can use this capability to validate uploaded documents without leaving the workflow, reducing the risk of incorrect file submissions and improving confidence when sharing attachments with customers.

Hierarchical Pick List Search Enhancements in Guided Workflow

This release enhances the search experience in Hierarchical Pick List components by improving search accuracy and ensuring that search results strictly adhere to builder configurations.

Previously, searches were limited to hierarchy levels such as L1, L2, and L3 and did not consistently include related entity fields configured as searchable. In addition, hierarchy records could be returned even when hierarchy levels were not marked as searchable, resulting in inconsistencies between configured settings and runtime behaviour.

With this enhancement, searches now include only the hierarchy levels and related entity fields that are explicitly marked as searchable. The search engine also respects the Searchable Hierarchies configuration, preventing hierarchy-level matches when no hierarchy levels have been configured for search. These improvements ensure that search results align with administrator-defined settings while maintaining existing auto-expansion and selection capabilities.

Workflow designers and administrators can use these enhancements to provide more relevant search results, improve data discoverability, and maintain greater control over how users search and select records within Hierarchical Pick List components.

Calendar Styling Enhancements for Live Chat Date Picker

This release introduces new styling options for the Date Picker calendar in Live Chat, allowing organisations to better control how the calendar is displayed within the chat interface.

Previously, the calendar layout offered limited flexibility, making it difficult to optimise the viewing experience for different Live Chat window sizes. The calendar could not be easily adjusted to utilise the available horizontal space or positioned according to design requirements.

With this enhancement, administrators can configure a Calendar Width setting to control the horizontal size of the calendar and create a wider, more user-friendly calendar view without requiring additional vertical space. A new Calendar Position setting also enables the calendar to be aligned left, centre, or right within the Live Chat window. By default, the calendar remains centre-aligned to provide a balanced user experience.

These enhancements give administrators greater control over Live Chat styling and help create a more responsive and visually consistent date-selection experience for end users.

Knowledge Base

The following features are being introduced in Sprinklr Service’s Knowledge Base module:

Unified Reporting Under Service Analytics

The following Knowledge Base Reports are now available under the Service Analytics Data Source. This change provides a single, unified reporting experience, improves dashboard performance, and ensures better data accuracy for real-time analytics. Refer to this page for more details on Knowledge Base Reporting.

The following reports are now grouped under the Knowledge Base Report Group in the Service Analytics Data Source:

• Knowledge Base Content Report

• Knowledge Base Language Variant Report

• Knowledge Base Usage Report.

Live Chat

The following features are being introduced in Sprinklr Service’s Live Chat module:

Live Chat A/B Testing

Live Chat now supports A/B experiments, enabling brands to make evidence‑based decisions about customer engagement. Brands can form a hypothesis (pick a goal metric), create an experiment with multiple variants that differ in UI, messaging, or behavior, and split audiences across those variants.

Brands can then analyze how each variant performs against chosen goals (whether that's engagement, conversions, or satisfaction) and compare the results. Once a clear winner is identified, they can roll out the winning variant or continue refining based on your findings. Experiments can also be targeted to specific audiences using URL, device type, geolocation, and authentication state.

Example Hypotheses Brands Can Explore

• Welcome message: Does a product‑focused greeting on product description pages outperform a conversational greeting leading to higher sales conversions?

• Trigger icon: Which icon style or color attracts more clicks?

• Home screen: Does reordering cards or changing descriptions increase self‑service usage?

• Conversation screen: Will modifying the screen title for a specific geolocation improve clarity and increase NPS?

Bot Message Feedback Form Theme Customization

The feedback form triggered by a thumbs‑down response in Live Chat now supports theme customization. Text elements within the feedback modal can be styled to match the brand’s theme, ensuring visual consistency across the conversation and improving readability.

Configurable elements include the header at the top of the form, the sub‑text below the header, the primary survey question, and the radio button option labels for selectable responses.

​

For further details, see Submitting Negative Feedback.

Zero‑Rated Proxy Support

Live Chat now supports routing traffic through client‑dedicated domains using CNAME mapping. With this enhancement, Live Chat sessions appear to originate from the brand’s own domain rather than Sprinklr‑owned domains.

Aligning traffic with the brand’s registered domain enables telecom operators to recognize interactions as part of zero‑rating agreements in metered mobile data markets. This means end‑users engaging with support do not consume their mobile data allowance when interacting with a zero‑rated brand. For example, when a customer in a metered data market opens a chat session, all messages and file uploads are delivered through the brand’s domain. The telecom operator identifies this traffic as zero‑rated, allowing the customer to continue the conversation without incurring data charges.

This capability removes a significant adoption barrier in data‑sensitive markets and provides organizations with a secure, compliant way to deliver cost‑free support experiences to their customers.

​

Messaging

The following features are being introduced in Sprinklr Service’s Messaging module:

WhatsApp Business - Request Contact Info Templates

The WhatsApp Business channel now supports two new capabilities: the Request Contact Info button in HSM Utility and Marketing templates, and a new Request Contact Info template type.

With these enhancements, brands can directly request contact information from WhatsApp users. This is particularly valuable when users are identified only by their BSUID/username and their phone numbers are not visible in the UI.

When a user shares their contact details, the phone number is ingested into Sprinklr, and the associated Sprinklr profile is updated. Profiles containing both BSUID and phone number coexist correctly across publishing and reporting, ensuring consistency across workflows.

For further details, see HSM - Request Contact Info and Request Contact Info Template.

Infobip SMS - Ability to Add Multiple Infobip SMS Accounts with Shared Sender ID

The Infobip channel now supports adding multiple SMS accounts with the same Sender ID. The account identifier has been updated to use API key and Project URL instead of Sender ID alone, allowing each account to be created and managed independently.

This improvement enables brands to configure several accounts that have a single Sender ID while separately managing them in Sprinklr. Brands can maintain a consistent channel identity across regions, comply with local regulations, and preserve account‑level visibility without risk of overwriting.

For further details, see Add an Infobip SMS Account.

WhatsApp Voice Message Transcription Reporting

You can now create reports in Sprinklr to gain visibility into transcription performance for WhatsApp voice messages using the data source: Inbound Analytics. Supported dimensions and metrics include detected language, audio duration, end‑to‑end latency, ML latency, and transcription status.

This enhancement provides clear operational monitoring, strengthens accuracy and SLA governance, and delivers transparency both across markets and within individual accounts.

For further details, see WhatsApp Voice Message Transcription Report.

Google RBM - Quick Reply Postback Capture and Exposure

For Google RBM messages containing Quick Reply buttons, the configured postback message is now captured and included in the inbound payload. When a user clicks a Quick Reply button, the postback data is recorded and made available for downstream use. This data can be leveraged in Rules, Bots, Guided Workflows, reporting, and other automation flows, enabling more precise handling of customer interactions.

Outbound Voice & Dialers

The following features are being introduced in Sprinklr Service’s Outbound Voice & Dialers module:

Support SSH Key-Based Authentication Without Requiring a Password

This release enhances authentication flexibility by allowing users to authenticate using either a password or an SSH key. Previously, the system required a password for authentication, even in scenarios where SSH-based authentication was being used. This created unnecessary friction for users who rely on SSH keys as their preferred and more secure authentication method.

With this update, users can now provide a password, an SSH key, or both when configuring authentication. At least one authentication method is required, but neither field is mandatory on its own. This change enables organisations to adopt SSH-based authentication workflows without being forced to configure a password.

The user interface has been updated to reflect this flexibility. The password field is no longer marked as mandatory, and users are guided with contextual messaging indicating that either a password or an SSH key must be provided. If neither field is supplied, the system displays an error prompting users to provide at least one authentication method.

Backend validation logic has also been updated to align with the new authentication requirements. Requests that do not contain either a password or an SSH key are rejected, while configurations containing a password, an SSH key, or both are accepted. Existing password-based configurations remain fully supported, ensuring backward compatibility without requiring any migration or reconfiguration.

This enhancement benefits administrators and users who manage integrations, servers, or systems that support SSH-based authentication. It can be used when creating new configurations or updating existing ones where password less authentication is preferred for security and operational efficiency. For further details, see ​Connector file Description.

Quality Management

The following features are being introduced in Sprinklr Service’s Quality Management module:

iFrame Widget Support in Case Analytics View

You can now add the standard iFrame widget to Case Analytics record page layouts. This enhancement allows administrators to embed external websites, tools, and custom applications directly within analytics pages, creating a more seamless experience for users. 

You can configure all supported iFrame widget settings through: 

• Settings panel 
• JSON editor 

This offers the following key benefits: 

Improve workflow continuity by providing access to external applications within analytics pages. 

Reduce context switching for users. 

Maintain a consistent configuration and administration experience across product areas. 

For more information, refer to Add and Configure Widgets in Case Analytics View.

Sprinklr AI Agent

The following features are being introduced in the Sprinklr AI Agent module:

Improved Guardrail Workflow Configuration

The Guardrails experience has been enhanced to make configuring and managing guardrail breach workflows more intuitive.

What's New

• The Replace breach response message with a custom workflow option is now accessible directly from the collapsed guardrail card.
• Custom workflows can be configured to replace the default guardrail breach response.
• A dedicated Guardrail Breach Workflow section displays all guardrails associated with a custom workflow.
• Guidance is displayed when no guardrails are configured for a custom workflow.

AI Agent Description Included in Task Prompts

AI Agent descriptions are now automatically included in task prompts sent to the underlying language model, helping generate responses that align more closely with the configured persona.

Enhanced RAG Reporting

Reporting for Retrieval-Augmented Generation (RAG) has been enhanced to provide deeper insights into response generation, knowledge retrieval, and guardrail execution.

What's New

• Reports now include User Message, Reworded Question, Generated Answer, Message ID, Case Number, and Engine ID.
• Added Response Grounded to indicate whether a response was generated using retrieved knowledge.
• Reports now display the knowledge articles used during response generation.
• Added Output Guardrail Time (ms) to measure the time taken to evaluate generated responses.

Simplified Task as a Tool Configuration

The Task as a Tool experience has been simplified to reduce configuration effort and improve task orchestration.

What's New

• Conversation summary prompt configurations are no longer required.
• You only need to select the target task to invoke.
• Child tasks automatically take control of the conversation and return control to the parent task after completion.
• Existing configurations continue to work without modification.

Chain Tool Calls from the End Dialogue Tree Node

You can now configure a follow-up tool directly within an End Dialogue Tree node when using Dialogue Trees as Tools.

What's New

• Added an optional Follow-up Tool field to the End Dialogue Tree node.
• Configured tools are invoked directly without requiring an intermediate model decision.
• Existing Dialogue Tree configurations continue to function without changes.
• The Follow-up Tool field can be left unconfigured when no additional action is required.

​

Support Entry Tool Execution for Tasks

Tasks can now execute an Entry Tool automatically before prompt execution begins.

What's New

• Added an optional Entry Tool setting in task configuration.
• The configured Entry Tool runs each time the task is invoked.
• Output generated by the Entry Tool is automatically added to the task context.
• Task execution begins only after the Entry Tool completes successfully.
• Tasks without an Entry Tool continue to follow the existing execution flow.

Expanded Dialogue Tree Node Availability

Additional Dialogue Tree nodes are now available across supported AI Agent workflows, enabling more flexible conversation orchestration.

What's New

• Additional nodes are now available, including Send Survey, Route to Another Bot, End Conversation, Schedule Callback, Merge Profile, Agent Handover, Deflect, Appointment Management, Typing Indicator, Streaming API, and AI Prompt nodes.
• These nodes can be used across Dialogue Tree tasks, Dialogue Tree tools, router tasks, conversation controls, and guardrail workflows.
• Existing configurations continue to function without modification.

Source-Agnostic Send Message Support

The Send Message node now supports Source Agnostic channels across Dialogue Tree tasks and Dialogue Tree tools.

Benefit

• Provides a consistent messaging experience across supported channel types.
• Ensures feature parity between the Send Message node and the Send Message tool.

Unified Evaluate Experience

A new Evaluate workspace brings evaluation capabilities together in a single experience.

What's New

• Evaluation Metrics, Scenarios, and Simulations are now available in one centralised location.
• Provides a streamlined workflow for evaluating and improving AI Agent performance.

Evaluation Metrics Management

Evaluation Metrics provides a centralised location to configure and manage AI Agent evaluation criteria.

What's New

• Standard conversation quality metrics are available by default.
• Added Containment to measure how effectively conversations are resolved without agent escalation.
• Added CSAT under Customer Experience metrics.
• Create custom evaluation metrics tailored to organisational requirements.
• Update metric configurations, evaluation questions, and supporting criteria as needed.

Dedicated Simulations Experience

A dedicated Simulations workspace is now available under Evaluate.

What's New

• Run simulations directly from scenarios and scenario sets.
• Monitor, filter, rerun, and review simulations from a centralised workspace.
• Supports streaming responses during simulation execution.
• Includes a Quality Score widget for reviewing evaluation outcomes.

Enhanced Scenario Generation

Scenario generation has been enhanced to support broader and more comprehensive AI Agent evaluations.

What's New

• Generate evaluation scenarios in bulk using AI Agent tasks, knowledge resources, guardrails, and historical conversations.
• Create custom evaluation scenarios using natural language descriptions.
• Automatically generate customer personas, scenarios, strategies, and success criteria.
• Edit or remove generated scenarios before evaluation.
• Supports streaming responses during scenario generation.

Enhanced Simulation and Evaluation Experience

The Simulation and Evaluation experience has been enhanced to provide deeper visibility into AI Agent performance.

What's New

• Simulation Summary provides consolidated performance insights across evaluation categories.
• View Scenarios includes coverage and quality metrics for generated scenarios.
• Audit Logs provide detailed simulation execution traces.
• An integrated Quality Score widget offers visibility into conversation quality and evaluation outcomes.

Added Support for Conversational Fillers

Voicebots now support conversational fillers to create a more natural customer experience during background processing.

Benefits

• Reduces perceived wait times while tools execute or information is retrieved.
• Plays contextual filler responses during background processing.
• Skips filler messages when the final response becomes available early.
• Creates a more natural conversational experience.

Reset Conversations Using Keywords

AI Agent Builder now supports resetting ongoing conversations using predefined keywords configured in Conversation Settings.

What's New

• Configure keywords that clear the current conversation and start a new one.
• Configured keywords are treated as the first message of a new conversation.
• Default keywords include Reset and Hi.

Sprinklr VoiceConnect

The following features are being introduced in the Sprinklr VoiceConnect module:

Outbound SIP header and Number Manipulation Policy

This release introduces a dedicated configuration module within the VoiceConnect Console for managing Outbound SIP Header Manipulation and Number Manipulation through Rules. These rules enables the user to add, remove, modify, or normalise SIP headers, as well as transform calling and called numbers to meet carrier-specific requirements and interoperability standards.

Previously, header and number manipulation configurations could only be implemented through backend. This approach increased operational overhead, required developer involvement, made auditing difficult, and slowed carrier onboarding and configuration changes. The new UI-based experience provides a centralised and more intuitive way to manage these rules directly from the VoiceConnect Console.

Users can create SIP header manipulation rules, configure number transformations such as prefix and suffix additions, and apply these configurations at the trunk level. Rules are managed independently and can be attached to relevant trunks as needed, making it easier to maintain carrier-specific configurations across telephony deployments. Validation checks are also performed before rules are applied, helping prevent malformed SIP messages and reducing the risk of call-processing issues.

Voice administrators, telecom engineers, implementation consultants, and operations teams can use this capability when onboarding carriers, configuring SIP interoperability requirements, managing dialling plans, or implementing network-specific routing logic. By reducing dependency on API-based changes and SBC-level updates, this enhancement improves configuration accuracy, governance, and operational efficiency. For further details, see Outbound SIP Header Manipulation Policy.

Unified Routing

The following features are being introduced in Sprinklr Service’s Unified Routing module:

Smart Assignment Reporting and Agent Score Analysis

This release enhances reporting capabilities by extending support for the Used Smart Assignment dimension across key analytics reports and business outcome metrics. These improvements help teams accurately measure the impact of Smart Assignment on case handling efficiency, resolution outcomes, and SLA performance, while also providing greater visibility into agent scoring behaviour.

Additionally, the functionality of the Used Smart Assignment dimension has been corrected in the Case SLA Report and Inbound Case Report to ensure accurate reporting. Support for the Used Smart Assignment dimension has also been added to the Case Processing SLA Report, creating a consistent reporting experience across SLA-focused analytics.

This release also introduces support for analysing Agent vs Score data based on the version of the Agent Score job used to generate live agent scores. This allows teams to compare scoring outcomes across versions and aligns score-version visibility with the Case × Agent Rank report. This enhancement applies to live score data and historical score records.

Supervisors, operations managers, business analysts, and reporting teams can use these enhancements to evaluate Smart Assignment effectiveness, monitor SLA compliance, analyse case distribution, and investigate agent scoring results with greater accuracy and confidence.

Use these reporting enhancements when assessing the business impact of Smart Assignment, reviewing operational performance metrics, troubleshooting SLA reporting discrepancies, comparing agent score versions, or validating agent ranking outcomes in live environments. For further details, see Smart Routing Reporting.

Add Agent and Agent-Level User Properties to Queue Performance Reports

This release enhances queue performance reporting by introducing Agent, Agent ID, and associated User Properties as reporting dimensions in the Voice Queue Performance, Digital Queue Performance, and Omni-channel Queue Performance reports. These additions provide more granular visibility into contact assignment data, enabling organisations to analyse queue distribution and performance using agent-level attributes captured at the time of assignment.

Many organisations operate multiple sites, product categories, or other business dimensions within the same work queue and need to understand how contacts are distributed across these dimensions. Previously, these attributes were primarily available at the case level, where values accumulated during transfers and often appeared as cross-multiplied or comma-separated data, making analysis difficult and limiting actionable insights.

With this enhancement, reporting can now leverage the agent associated with the assignment event, along with the agent's corresponding user properties, to provide a more accurate representation of contact distribution. This enables teams to analyse assignment patterns and operational performance based on business-specific agent attributes such as customer centre, product category, and other custom user properties.

The new Agent field displays the Sprinklr user associated with the contact assignment used for queue performance calculations such as Assigned, Answered, and related metrics. The Agent ID field provides the corresponding user identifier for the same assignment. In scenarios involving queue-to-queue transfers, a new contact record is generated and reported as a separate row. For agent-to-agent transfers, no additional contact record is created, ensuring reporting consistency.

This release also extends support for Agent User Properties, allowing organisations to report on custom user fields associated with the assigned agent. Both current-value and snapshot-value reporting scenarios are supported, helping teams preserve accurate assignment context for operational analysis and historical reporting.

Supervisors, operations managers, workforce management teams, and business analysts can use these enhancements to gain deeper visibility into contact routing, evaluate distribution strategies, analyse performance across product lines, and generate more meaningful queue performance insights.

Use these capabilities when monitoring assignment distribution across business dimensions, validating percentage-based routing strategies, analysing agent workload allocation, or reporting on queue performance using agent-specific attributes that are captured at the time of assignment. For further details, see Omnichannel Queue Report.

​Prevent Configuration Overwrites with Work Queue Version Control

This release introduces version control and collaboration awareness for Work Queue configuration management, helping administrators and operations teams safely manage concurrent updates. The enhancement prevents users from unintentionally overwriting changes made by others.

Previously, when multiple users edited the same work queue simultaneously, one user's changes could be overwritten by another user saving from an outdated editor session. Because users were not informed that a newer version had already been saved, this created a risk of data loss, configuration inconsistencies, and operational confusion.

With this enhancement, every work queue editor session tracks the version of the configuration that was loaded when the editor was opened. When a user saves changes, the system validates the editor version against the latest saved version. If another user has already saved a newer configuration, the system prevents the outdated version from silently overwriting those changes and notifies the user that their editor is no longer aligned with the latest version. Users are guided to refresh, reload, or reconcile their changes before attempting to save again. In addition, the editor now displays collaboration presence information, allowing users to see how many people currently have the same work queue open for editing.

Queue administrators, operations teams, and supervisors can use this capability when managing routing logic, assignment rules, and other work queue configurations in environments where multiple users may be making updates simultaneously. By preventing stale saves and improving collaboration awareness, the enhancement reduces the risk of configuration errors, protects recently saved changes, and enables more confident management of work queue updates. For further details, see Creating and Managing Work Queues.

Voice AI

Bring Your Own Key (BYOK) Support for TTS and STT

Sprinklr now enables administrators to configure and use customer-managed provider credentials for supported Voice AI providers through a self-service Bring Your Own Key (BYOK) experience.

Administrators can add provider credentials from AI+ Studio and use those providers across Text-to-Speech (TTS) and Speech-to-Text (STT) workflows.

Once configured, provider resources are automatically synchronized and become available within Voice AI experiences.

​

What’s New

• Configure supported Voice AI providers using your own provider credentials.
• Use BYOK-enabled providers in both TTS and STT configurations.
• View credential ownership through the new Key Management indicator.
• Access provider-specific voices after synchronization completes.

​

Notes

• Newly added providers may require up to two hours to synchronize.
• Resources configured through customer-managed credentials are identified with a BYOK label.

For further details, see Configure BYOK for Voice AI Providers.

Expanded Provider Support for TTS and STT

Sprinklr has expanded Voice AI provider support across Text-to-Speech (TTS) and Speech-to-Text (STT), giving
organizations greater flexibility when configuring speech generation and transcription workflows.

With this release, customers can use additional Voice AI providers directly within Voice AI configurations, enabling broader provider choice for both speech synthesis and transcription use cases.

The following provider support has been introduced:

​

Azure OpenAI Support

Azure OpenAI is now available as a supported provider for both Text-to-Speech and Speech-to-Text configurations.
Administrators can select Azure OpenAI when configuring speech generation and transcription workflows.

​

Cartesia Support for Text-to-Speech

Cartesia is now available as a supported Text-to-Speech provider. Customers can use Cartesia voices when creating and managing TTS configurations.

​

Sprinklr In-House Support for Speech-to-Text

Voice AI now supports Sprinklr In-House for Speech-to-Text workflows, providing an additional transcription option within STT provider configurations.

For further details, see Configure Voice AI TTS and Configure Voice AI STT.

Voice IVR

The following features are being introduced in Sprinklr Service’s Voice IVR module:

Expose Dead Air and Talk Time Metrics in Voice Conversations

This release introduces Dead Air Time, Talk Time, and Dead Air Percentage as voice conversation variables, providing greater visibility into how call time is distributed between active conversation and periods of silence. These metrics help organisations better understand call quality, customer engagement, and operational efficiency during voice interactions.

Customers often use dead air analysis to determine whether callers are experiencing excessive silence during conversations and to identify opportunities for process improvement. By making these metrics available directly within voice conversations, teams can evaluate the proportion of a call spent in active discussion versus idle time and use this information to drive workflows, routing decisions, reporting, and operational reviews.

With this enhancement, Talk Time displays the total duration of active conversation during the call, while Dead Air Time captures the amount of detected silence. The system also calculates Dead Air Percentage, allowing users to quickly assess the share of the overall interaction that consisted of dead air. These metrics can be leveraged in analytics and business processes requiring deeper visibility into conversation behaviour and customer experience.

Supervisors, quality analysts, operations teams, and contact centre administrators can use these insights to identify interactions with excessive silence, monitor call quality trends, and support performance improvement initiatives. For further details, see Dead Air and Talk Time Metrics in Voice Conversations.

Simplify IVR and Bot Audio Management with Dynamic File Name-Based Lookup

This release introduces dynamic file name-based media retrieval for IVR and Bot experiences, reducing the operational effort required to manage and update audio files. Instead of relying solely on asset IDs, IVR and Bot nodes can now locate and play audio files using file.

Previously, whenever audio files were refreshed or replaced in the media library, operations teams had to manually update the corresponding asset IDs across all affected IVR and Bot nodes. This process increased maintenance overhead, slowed routine content updates, and created opportunities for configuration mismatches.

With this enhancement, you can fetch audio with asset name from IVR and bot using static names or dynamic variables.

The lookup process supports several common scenarios. When a unique file with the specified name exists, the system retrieves and plays that file. If multiple files share the same name, the first matching file is selected, similar to record retrieval behaviour in other platform experiences. If no file matching the generated name is found, the node transitions to a failure state, allowing organisations to configure appropriate fallback or error-handling logic.

Operations teams, contact centre administrators, and IVR designers can use this capability to simplify media maintenance, automate audio selection based on call context, and reduce the effort required to manage frequently changing announcements, campaign messaging, language-specific prompts, or business-specific recordings.

Use this feature when audio files need to be dynamically selected based on customer, campaign, language, region, or other call attributes, or when frequent media updates make manual asset ID management difficult and time-consuming.

View Child IVR Flows Directly Within the Parent IVR Experience

This release enhances the IVR design and troubleshooting experience by allowing users to view child IVR processes directly within the parent IVR when a Call Another Flow node invokes a separate IVR. By providing visibility into connected IVR flows within the same workspace, teams can better understand end-to-end call journeys without navigating away from the current screen in a separate Sprinklr tab.

Previously, users had to open the child IVR in a separate browser tab to understand how the invoked flow was configured. This required frequent context switching between tabs, making it harder to follow call paths, analyse flow behaviour, and troubleshoot complex IVR implementations.

With this enhancement, the child IVR process is displayed inline within the parent IVR under the Call Another Flow section. Users can seamlessly explore downstream flows while retaining context of the parent journey, creating a more connected and intuitive design experience. In addition, node-level analytics from the child IVR are surfaced within this embedded view, providing deeper visibility into flow performance and customer progression across linked IVRs.

This improvement enables IVR designers, administrators, and operations teams to analyse complete call journeys from a single location, making it easier to understand how different IVR processes interact and contribute to the overall customer experience.

Teams can use this capability when reviewing complex IVR structures, investigating customer journeys that span multiple flows, validating call routing logic, or troubleshooting issues within interconnected IVR experiences. By eliminating unnecessary tab switching and providing analytics within the same context, the enhancement improves productivity and accelerates flow analysis.

Supervisors, contact centre administrators, IVR designers, and support teams benefit from a more efficient way to visualise, analyse, and maintain multi-level IVR implementations. For further details, see Viewing Child IVR Processes in Reporting.

Generate AI-Powered Summaries for IVR Flows

This release introduces AI-powered IVR summarisation, enabling users to quickly understand the purpose, structure, and customer experience of an IVR flow without inspecting individual nodes. By generating a plain-language overview on demand, the feature helps users comprehend complex IVR configurations within seconds and reduces the effort required to analyse or troubleshoot call flows.

Previously, understanding an IVR required manually reviewing nodes, transitions, conditions, and routing logic. This made it difficult to determine the business purpose of a flow, understand the customer journey, identify critical decision points, or evaluate the possible outcomes available to callers.

With this enhancement, users can generate a natural-language summary directly from the IVR experience. The summary explains what the IVR is designed to accomplish, how customers progress through the flow, the key interactions and routing decisions that influence the journey, and the possible endpoints customers may reach. This allows teams to gain an immediate understanding of an IVR's behaviour without needing detailed technical knowledge of the underlying configuration.

The generated summary includes a business-level overview of the IVR, a description of the customer journey, identification of important branching and decision points, and a consolidated view of the major outcomes available within the flow. These outcomes may include routing to an agent, transferring to another flow, directing callers to a queue, invoking fallback logic, or ending the interaction.

To support ongoing IVR maintenance, users can regenerate summaries after making changes to a flow, ensuring that the generated content remains aligned with the latest configuration. Summaries can also be copied and shared with stakeholders, making it easier to collaborate across implementation, support, and operational teams.

Adminstrators, implementation consultants, support teams, and contact centre administrators can use this capability to accelerate IVR reviews, simplify onboarding for new team members, improve troubleshooting efficiency, and facilitate discussions with business stakeholders who may not be familiar with IVR design details.

Use this feature when analysing existing IVR configurations, validating customer journeys, documenting call flows, troubleshooting routing behaviour, preparing knowledge-transfer materials, or explaining IVR functionality to internal teams and customers. By providing an instant business-level understanding of IVR logic, the feature reduces analysis time and improves overall visibility into IVR operations. For further details, see Generate IVR Summary Using Sprinklr AI.

Organise User-Created Resources with a Dedicated Custom Resources Section

This release improves resource management by introducing a dedicated Custom Resources section within the Managed Resources area. The new section provides a clear separation between platform-managed resources and user-created resources, making it easier for users to locate, manage, and maintain custom assets.

Previously, user-created resources were displayed alongside other managed resources, which could make navigation and resource discovery more difficult, especially in environments with a large number of resources. As organisations scale their implementations, identifying custom-created resources quickly becomes increasingly important for administration, maintenance, and governance activities.

With this enhancement, all user-created resources are grouped under a dedicated Custom Resources section. This provides a more organised management experience, improves visibility into customer-specific assets, and helps users distinguish between resources created by their teams and those provided or managed by the platform.

The new structure enables administrators and resource managers to access relevant resources more efficiently, reducing the time spent searching through mixed resource inventories. It also improves overall usability by creating a more intuitive and streamlined navigation experience within the Managed Resources interface.

Administrators, implementation teams, and operations users can leverage this enhancement when managing custom configurations, reviewing customer-specific assets, or maintaining resources created for unique business requirements.

Workforce Management

The following features are being introduced in Sprinklr Service’s Workforce Management module:

Enhanced Audit Trail for Agent Movements

When viewing an agent’s schedule audit for a Schedule Scenario, you can now clearly track agent transfers between scenarios:

• Agent Moved Out is recorded in the source Schedule Scenario when an agent is moved out.

• Agent Moved In is recorded in the destination Schedule Scenario when an agent is moved in.

This enhancement improves the visibility and traceability of agent movements across Schedule Scenarios.

Simplified Work Type Assignment During Agent Transfers

While moving agents between Schedule Scenarios, agents associated with Work Types that exist in both the source and destination scenarios are now automatically populated in the destination scenario based on their assignments in the source scenario. You can review and modify these assignments as needed, reducing manual effort and improving accuracy. This enhancement reduces manual effort and streamlines the agent transfer process.

Introduced New Permission to Control Mobile Status Changes

We have introduced the Allow Status Change permission to control whether agents can change their status from the Sprinklr Mobile Application. Only agents assigned this permission can update their status using the mobile app.

This enhancement helps prevent manual overrides of scheduled statuses, improving Real-Time Adherence (RTA) accuracy and maintaining the integrity of Workforce Management reporting. For further details, see Workforce Management Permissions.

Thank you for using Sprinklr Service. The 26.7.1 release brings significant enhancements and new features designed to make your experience even better. Please contact us at tickets@sprinklr.com for any questions or assistance.