Tableau Dashboard Actions: Building Interactive, Linked Analytics Experiences

Tableau dashboard actions are the mechanism for making dashboards interactive — transforming them from static views into linked, user-driven analytics experiences. An action is a rule that fires when a user performs an interaction (click, select, hover, or menu selection) on a source sheet, and changes something in the dashboard in response: applies a filter to another sheet, changes a parameter value, updates a set, highlights marks, or opens a URL. Understanding the full range of action types and how to combine them is the foundation for building Tableau dashboards that analysts and business users want to use repeatedly.

Action Types

**Filter actions** — the most common action type. When a user selects marks in a source sheet, filter actions apply a filter to one or more target sheets based on the selected dimension values. The target sheets update to show only data for the selected context. When the user deselects, the filters can clear (showing all data), keep the last selection, or exclude the selected values.

Filter action run behaviour options:

- *Run on select* — fires when marks are selected by clicking

- *Run on hover* — fires as the user mouses over marks, useful for exploratory scenarios

- *Run on menu* — adds a named action to the right-click context menu, keeping the dashboard clean until the user explicitly chooses the action

**Highlight actions** — visually emphasise related marks across multiple views without filtering. When a user selects marks in a source sheet, matching marks in target sheets are highlighted; non-matching marks are dimmed. The data is not filtered — all marks remain visible. Highlight actions are appropriate when the user needs context from the full population while focusing attention on the selected subset.

**URL actions** — open a URL when a user interacts with a mark. The URL can be static or dynamic, incorporating field values from the selected mark using field name tokens in the URL template. Common uses: linking to a CRM record for a selected customer, opening a documentation page for a selected product code, navigating to a detailed report for a selected region.

Dynamic URL example: https://crm.internal/customers/[Customer ID] — where [Customer ID] is replaced at runtime with the field value of the selected mark. URL actions can open in the current browser window, a new tab, or a web object on the dashboard.

**Set actions** — update set membership based on user interaction. When a user selects marks, the dimension values of those marks are added to, removed from, or used to replace the membership of a target set. Any calculations and views that reference the set update immediately. Set actions are the mechanism for portfolio builder dashboards, custom peer group selection, and any pattern where the user defines the analytical group interactively.

**Parameter actions** — update a parameter value based on user interaction. When a user clicks or hovers on a mark, a parameter action sets the parameter to a value from that mark's data. Any calculations and views that reference the parameter update immediately. Parameter actions are used to drive single-selection comparative patterns — clicking a customer to show that customer's trend highlighted against the population.

Filter Action Design Patterns

The most important design decision for filter actions is the run-on-clear behaviour:

**Clear filter** — when the user deselects, the target sheets show all data. This is the default and appropriate for most drill-down patterns.

**Keep filter values** — the last selected filter remains active even after deselection. Appropriate when the user is making an explicit "I want to focus on this" selection that should persist until they select something different.

**Exclude all values** — when the user deselects, the target sheets show no data. This is rarely useful for general-purpose dashboards but is occasionally appropriate for side-by-side comparison patterns where the detail panel should be blank until a selection is made.

Filter Actions vs Parameter Actions: When to Use Each

Filter actions and parameter actions are both used to drive cross-sheet context changes, but they behave differently in important ways:

**Filter actions** filter the target sheet's data — they affect which rows are included in the aggregation. They can handle multi-mark selections (the user selects 3 customers, the target sheet shows data for all 3). They reset naturally when deselected.

**Parameter actions** change a parameter value — they affect calculations that reference the parameter. They only work with single-value selections (a parameter holds one value; selecting multiple marks sets the parameter to the value from the last mark). The parameter state persists after deselection.

Use filter actions for most drill-down patterns. Use parameter actions when the selected context needs to drive calculations (not just filtering), reference lines, or cross-datasource behaviour, and when the selection is always single-valued.

Building a Master-Detail Dashboard with Filter Actions

The master-detail pattern — a summary list on one side, detail view on the other — is one of the most common filter action use cases:

1. Create a summary view (bar chart, table) showing all entities with a key metric.

2. Create a detail view (line chart, text table) that will show detail for the selected entity.

3. Create a filter action: source = summary view, target = detail view, run on select, filter on the entity dimension, clear on deselect.

4. On load, the detail view shows all data (the "no selection" state). When the user clicks an entity in the summary, the detail view filters to that entity.

For a cleaner user experience, add an instruction tooltip to the summary view ("Click to see detail") and a title to the detail view that updates via a calculation when a selection is made.

Tooltip Actions as Discovery Surfaces

A tooltip action is a URL action or sheet tooltip configured to provide contextual information when the user hovers. This keeps the main dashboard clean while making supplementary information available on demand.

For operational dashboards: hovering over a KPI card opens a tooltip showing the trend for that KPI over the last 30 days (via Viz in Tooltip). The user gets the trend context without a secondary chart permanently on the dashboard taking up space for users who do not need it.

For data governance dashboards: hovering over a data quality indicator opens a tooltip showing the specific row-level issues driving the quality score.

Action Sequencing and Cascades

Multiple actions can be chained. A filter action on Sheet A updates Sheet B; a separate filter action on Sheet B updates Sheet C. The result is a cascading drill-down where each selection at one level of granularity drives the next level.

Cascade depth should be limited to three levels. Deeper cascades become difficult for users to follow and tend to produce low-data views at the bottom of the chain. When the analytical goal requires more than three levels, consider navigating to a separate dashboard at the deeper level rather than adding more views to the existing dashboard.

Performance Considerations

Actions add query overhead: each action fires a query to re-render the target sheets. For dashboards with multiple target sheets and large data sources, action performance can be slow. Optimisation approaches:

**Limit target sheet scope** — only include sheets as filter action targets that actually need to respond to the interaction. Avoid "all sheets on dashboard" as the target for filter actions on large dashboards.

**Use extracts** — actions on live connections query the database on every interaction; actions on extracts query the in-memory extract engine. For action-heavy dashboards, extract-based data sources deliver significantly better interactivity.

**Pre-aggregate** — if the detail view shows pre-aggregated data (not row-level), the query returning the filtered result is faster than a row-level query. Design data sources to match the granularity that the detail view actually needs.

Our Tableau consulting practice designs interactive Tableau dashboard experiences for enterprise clients — contact us to discuss dashboard interaction design for your analytics environment.

A former Microsoft data architect audits your data foundation, identifies your top priorities, and sends you a written plan. Free. No pitch.