Assistant

Last updated: October 5, 2026

The assistant is an AI-first interface for end-to-end analytics creation, from data acquisition to data modeling, insights generation, and embedding. The assistant also provides a conversational interface available directly within Sisense dashboards, designed to empower users with self-service analytics through natural language interaction. The assistant is available for managed cloud customers, as well as self-hosted customers (as long as the prerequisites are met). It supports data designers, dashboard designers, and end users in quickly creating analytics, gaining insights, exploring data, and building visualizations directly within the dashboard environment. This feature will ultimately replace our more traditional natural language query feature, Simply Ask, offering a modern, intuitive experience aligned with the expectations of today’s data consumers. By simplifying analytics and widget creation and exploration, the assistant addresses core usability and productivity challenges faced by data designers, dashboard designers, and data viewers.

Depending on where you are in the workflow, the assistant appears in different contexts with tailored capabilities:

• Studio assistant: Available in a dedicated tab, designed for creators working across the full analytics lifecycle: from data modeling, to chart and dashboard creation, to embedding.

• Assistant sidebar: Available in the Analytics tab (dashboards). The assistant sidebar acts as a companion to dashboards, enabling users to ask questions and explore data while interacting with the dashboards and widgets. If the user can edit the current dashboard, the assistant can also add new widgets to the dashboard to save AI-generated insights. The assistant sidebar is also available for dashboards embedded via iFrame, extending AI capabilities to any audience (see Embedding the Assistant).

Use Cases and Benefits

Use Cases

• Data Designers can create analytics, from data acquisition to data modeling, generate insights, and embed.

• Dashboard Owners have access to the assistant if it is enabled for them (see Enabling the Assistant) and the data sources allow it. They can converse with the assistant to get insights on the data while creating, refining, and adding new widgets to dashboards.

• Users with Design Permissions: Any user the dashboard is shared with who has design permissions (including Viewer Plus) can also create and add widgets through the assistant.

• Users with View Permissions: These users can use the assistant to ask questions and generate insights conversationally, including asking the assistant to create widgets based on the data. They cannot add widgets to a dashboard they have view permissions for, though any widgets they create remain in the conversation history for future reference.

Benefits

The assistant democratizes access to data by allowing users to interact with dashboards through natural language, reducing reliance on technical teams for generating insights. It accelerates dashboard design and iteration, making it easier for designers to create and refine content in real-time. The assistant supports both open-ended data exploration, where users ask spontaneous questions, and more structured, guided analysis, helping users uncover insights in ways that traditional dashboards alone cannot.

Supported LLMs

The assistant requires an LLM and Cloud Linked Features must be enabled. This applies to both managed cloud and self-hosted deployments. You can use the Sisense managed LLM, where Sisense provides and operates the model and no configuration is required, or bring your own LLM and connect your own provider. For the current list of supported providers and configuration steps, see Setting Up Your LLM.

Enabling the Assistant

In order to use the assistant, it must be enabled by an Admin. The assistant is managed through centralized AI feature management in the Admin tab, under Sisense Intelligence, which uses a cascading access control model:

• General default — the baseline AI access that all tenants inherit, including the global master toggle for cloud-linked features. The assistant is a cloud-linked feature, so Cloud Linked Features must be enabled.

• Tenant level — per-tenant overrides of the general default.

• User group level — per-group permissions within a tenant.

Lower levels inherit from the level above and cannot exceed it.

1. Enable the assistant at the tenant level in the Admin tab, under Sisense Intelligence > Enablement.

   

2. Enable the assistant at the user groups level in the Admin tab, under Sisense Intelligence > Feature Management.

   

Dashboard owners must then activate the assistant individually for each dashboard, using the > Share the assistant toggle in the dashboard menu, before sharing or republishing the dashboard to the users who should have access to the assistant. Alternatively, if the Enable for all users option is selected on the admin page, the toggle is enabled by default; the owner can disable it before sharing or republishing the dashboard to prevent recipients from using the assistant.

Whenever this toggle is modified, a confirmation dialog explains the impact and ensures that users only get access to the assistant as intended.

Whenever the assistant sidebar is available but not visible, users see the floating assistant button over the bottom-right corner of the dashboard. Clicking the button opens the assistant sidebar; closing the sidebar makes the button reappear.

For more information about enabling Generative AI and setting up LLM access, refer to the Enabling Generative AI and Setting Up Your LLM documentation.

Admins can allocate monthly credit budgets per tenant and user group, and monitor AI credit consumption by tenant, user group, and feature. See AI Credit Usage and Allocation. When a tenant's or user group's credits are used up, the assistant is unavailable to its users until the credits reset or more are allocated.

Optional Enhancements

While the following enhancements are optional, they are recommended, as a clean, structured model provides better clarity and improves the quality of insights. See Optimizing Your Data Model for AI for more information.

Using the Assistant in Dashboards

To use the assistant, click the floating assistant button .

In the assistant, you can:

• Type your question in natural language

• Ask the assistant to describe the current dashboard or widget

• Choose one of the suggested starter buttons

Active Context Chips

The assistant sidebar knows which dashboard or widget is currently on screen, so it can respond effectively to requests such as "What am I looking at?" or "Describe this."

When a dashboard is active, the dashboard name is displayed above the text input area.

When a widget is opened in the widget editor or in the full-screen view available to viewers, the widget name is also displayed.

This context helps the assistant better understand users' intentions and provide better responses.

Types of Questions Supported

The assistant can respond to questions about itself, your data model, your data, and the current dashboard or widget.

• Ask "Who are you?" or "How can you help me?" to learn what the assistant can do.

• Ask "How many tables are in my data model?" or "Which table has customer status?" to explore your data model.

• Ask analytical questions, such as "How many active customers do we have?" or "Top five regions by revenue" to generate insights and visualizations.

• Ask exploratory analytical questions, such as "What caused the spike in September?" These can trigger additional autonomous queries in which the assistant tries to find key insights in the data and explain them in natural language.

Before presenting results, the assistant shows how it interpreted your question as a set of chips covering the measures, dimensions, and filters it used, along with any dashboard filters applied.

Supported Natural Language Query Capabilities

The assistant supports common analytical queries, enabling users to apply familiar functions and filters in natural language.

Formulas

• Aggregations: sum, count, countDistinct, min, max, median, average

• Quick functions: contribution, pastYear, difference

• Arithmetic operations using words (e.g., "revenue minus cost") or symbols (e.g., revenue - cost)

Filters

• MemberFilter: Filter by specific values

• FromToFilter: Filter values in a range

• ValueFilter: fromNotEqual, from, toNotEqual, to, equals, doesntEqual, top, bottom

• ExcludeFilter: Exclude specific values

• LastFilter / NextFilter: Time-based filters (e.g., "last month", "next year")

• Contains / Doesn’t Contain: Filter string values containing/excluding text

• StartsWith / EndsWith: String matching from the start or end

• ByTopFilter / ByBottomFilter: Top or bottom values using a metric

Visualizations

Users can request specific chart types or let the assistant choose the most appropriate one.

Supported visualizations include: Area chart, Bar chart, Column chart, Funnel chart, Indicator chart, Line chart, Pie chart, Polar chart, Table, and Treemap chart.

Results

The Results pane offers the following options:

• Switch to Table (): Show the data results in a table instead of a chart

• Download (): Export the results as an image or CSV file

• Add to Dashboard (): For users with design permissions, this option allows you to add the generated chart as a widget directly to the dashboard

Limitations and Known Issues

The assistant is evolving rapidly. We are continuously improving its capabilities both proactively and based on user feedback. As a result, some current limitations may be addressed in future releases.

Natural Language Query (NLQ) Limitations

• No access to historical user queries: Sisense does not currently provide visibility into what NLQ questions were previously asked by users. This limits auditing, usage insights, and optimization efforts.

• Cannot create persistent calculated fields: The assistant can perform arithmetic across fields within a query, such as "revenue minus cost", and can use calculated dimensions already defined in the data model. It cannot create a new calculated field or dimension and save it to the model. Reusable metrics must be defined in the data model or as shared formulas.

• Limited real-time learning from feedback: Thumbs up/down feedback is collected but does not immediately influence the assistant’s behavior.

Semantic Layer and Smart Matching Limitations

• Semantic meaning required: Does not support numerical codes or abbreviations that lack semantic meaning.

• No direct synonym mapping: You cannot define explicit synonym pairs or override how a term is matched. Synonym detection is handled by the LLM. You can influence it indirectly by providing clear field and table descriptions, adding business terminology through AI Context, and enabling AI Smart Value Matching.

• Limited support for ambiguous or poorly labeled fields: Vague field names like "value1" or "temp_field" can reduce interpretation accuracy.

• Column names with periods (.) may cause errors: Avoid using “.” in field names.

Dashboard Awareness and Workflow Limitations

• Modification of existing widgets or dashboards: The assistant is aware of the widgets on the active dashboard, but it cannot currently modify those widgets or the dashboard itself. However, it can create new widgets based on the definition of an existing one.

• Partial awareness of existing formulas: The assistant is aware only of shared formulas. It is not aware of other formulas in the dashboard which are not shared.

Presentation and Styling Limitations

• Limited styling options: Generated visuals may not match your dashboard’s formatting.

• Manual adjustments often required: Layout, labels, and formatting may need to be customized post-creation.

Data Model Limitations

• Schema changes: In order for schema changes to the data model to take effect, you must refresh the chat session.

Time Limitations

• Timestamp support: The assistant only supports timestamps in units of days or longer (i.e., it does not support hours or minutes).

Stay up to date with the latest product release notes and documentation to track feature improvements and known limitations.

Embedding the Assistant

The assistant sidebar can be embedded into external applications alongside dashboards using iFrame embedding, giving any audience access to the conversational interface.

To enable the assistant sidebar for an iFrame-embedded dashboard, add the a=true parameter to the dashboard URL (for example, &a=true). The parameter defaults to false; when it is set to true, the floating assistant button appears on the embedded dashboard. The user must also be authorized to access the assistant.

For more details, see Available Parameters in the Developer documentation.

Creating Analytics with the Assistant

To use the assistant to create analytics, open the Assistant tab.

In the assistant, you can:

• Type your question in natural language

• Choose from suggested questions

• Create a new chat session and navigate between sessions

• Upload a file (pdf or image) for additional context (experimental)

The following table depicts what you can do with the assistant:

	Created by the assistant	Display/Edit from Fusion	Actions
Data Model	Create a data model with synthetic data or csv uploads	N/A	Save and build
Chart	Create and Modify chart attributes	Once added to a dashboard, edit in the Widget Editor with the appropriate permissions	Get embed code
Dashboard	Create and Modify a Dashboard*	Display and modify with the appropriate permissions, as with any dashboard	Get embed code

* Note that the assistant respects data governance rules defined in the product, and therefore you can only create dashboards on data sources with the appropriate permissions (can `use` or `edit`).

Creating Data Models from the Assistant

You can ask the assistant to create a data model:

1. For example, in the Assistant tab, choose the Create a data model option.

   

   The data model is generated and the tables to be included are displayed.

   

2. Review the tables to be included. Either manually or conversationally, you can add additional tables, edit the existing tables, or leave everything as-is.

   • To edit an existing table, expand the table via the drop-down, and edit the fields as desired. Note that in addition to the text/numeric fields, there are also "Faker options" fields. These are used to apply relevant constraints, such as setting date ranges, numeric limits, or custom formats, to the generated column values based on the column name.

     

   Note:

   The synthetic data (generated data, not real-world data) may hold content that does not match the semantic meaning of the column. This functionality provides the user with a quick way to prototype and design the data model and visualization, without paying attention to the data insights.

3. When you are ready, click Create Data Model to build the data model into an ElastiCube. When the process is complete, a “Data generated successfully” message is displayed.

   

4. Click the Data tab and select your newly created data model to review the complete structure.

   

   

Creating, Displaying, and Modifying Dashboards from the Assistant

When creating a new dashboard via the assistant, it is saved automatically. It can then be displayed and modified later, under the Analytics tab, with all other dashboards.

You can use the assistant to create a dashboard for you, either from a pre-existing data model, or even from the data model that you created using the assistant.

1. For example (after creating a data model as described above), in the Assistant tab, type “Create a dashboard based on this data model” and press Enter.

   You may be requested to select which data source to use. The assistant then generates the requested dashboard.

   

2. To save the dashboard for later viewing in the Analytics tab, click the Save button.

   

3. After saving it, to view the dashboard that you just saved, click Open Dashboard.

   

   The dashboard is then displayed in the Analytics tab.

4. Click to customize any widget in the Widget Editor, just as you would for any dashboard.

API Reference

Sisense offers API support for working with the assistant and its underlying AI capabilities.

• REST APIs — Support major workflows, including Generative AI setup, NLQ (Natural Language Query) execution, and completion APIs for text-based interactions.

• Compose SDK — Provides visual components and programmatic APIs for integrating standalone assistant functionality and embedding custom experiences within applications.

• Feature access control — Enable or disable AI features, including the assistant, at a given scope (PUT /ai/features/{scope}).

To learn more, see the Developer documentation.

Frequently Asked Questions (FAQ)

Q: Does the assistant support Viewers?

A: Yes, Viewers can use the assistant to ask questions and download results, but they cannot add or edit widgets. Viewer Plus users can also save assistant-generated widgets to dashboards they own or that are shared with them with Can Design permission.

Q: Can I customize responses or define synonyms?

A: No. Synonym matching is handled automatically through semantic inference. Manual mapping is not currently supported.

Q: Is this a replacement for Simply Ask?

A: Yes, the assistant is the next-generation natural language query experience, offering a more intuitive and flexible interface. As the assistant matures and becomes widely available to self hosted customers, we will communicate the plan for Simply Ask.

Q: Can the assistant access or reference existing widgets on a dashboard?

A: Yes, it can reference existing widgets to help generate new insights and charts, but it cannot interact with existing dashboard elements.

Q: Can I train the assistant on custom business terminology?

A: Not directly. The assistant relies on your data model's metadata, including column names and descriptions. Clear, business-friendly naming improves understanding, but there is no manual training or synonym configuration at this time.

Q: How can I improve the assistant’s accuracy on domain-specific queries?

A: Use clear field names, provide descriptive metadata, and enable AI Smart Matching where appropriate. For complex or calculated metrics, consider defining formulas in the model or using perspectives to reduce ambiguity.

Q: Does the assistant retain context between questions?

A: Yes. The assistant remembers previous questions within the same session, can reference the results of queries it has already run, and can run additional queries on its own to answer follow-up questions.