> AI agents: this is one page from Mammoth Analytics documentation. The index of all pages as Markdown is https://docs.mammoth.io/llms.txt. Append `.md` to any docs URL, or send `Accept: text/markdown`, to get Markdown.

# View management

> Master Views, Batches, and version control to organize data transformations and collaborate effectively with your team.

A **View** is your transformation workspace within a dataset. It starts with the complete combined data from all batches and applies a pipeline of tasks—filters, calculations, joins, reshaping—to produce a transformed output ready for analysis, reporting, or export.

Every dataset can have multiple Views. Each View has its own independent pipeline, its own settings, and its own output destinations—while all Views share the same underlying data. Your team can work on the same dataset at the same time without changing each other's pipelines.

![Data Library with a dataset expanded to View 1 and View 2 and an Overview panel showing 2 Views](https://docs.mammoth.io/api/v1/images/20260409_081844_Datasetoverviewshowingmultipleviews.jpg)

## Why use multiple Views in a dataset?

The same dataset rarely serves just one audience. A sales dataset might need a detailed analyst pipeline, a KPI summary for the operations team, a formatted feed for a dashboard, and a compliance export for finance—all from the same source, all maintained independently.

Without multiple Views, teams either duplicate data into separate files (creating maintenance burden and version drift) or share a single pipeline (creating conflicts and fragility). Multiple Views avoid both problems.

Multiple Views enable three things.

**Parallel team work.** Two analysts can build completely different pipelines on the same dataset simultaneously. Changes in one View never affect another.

**Audience-specific outputs.** Each stakeholder group gets data shaped precisely for their needs—without the underlying dataset being restructured or copied.

**Safe iteration.** Duplicate a working View to experiment with new logic. If the experiment fails, delete it. The original is untouched.

![Dataset open in the editor with three View tabs along the bottom and a data grid above](https://docs.mammoth.io/api/v1/images/20260409_082040_Onedatasetwiththreeviewtabs.jpg)

## Common View patterns

These are not rigid categories—they are patterns that emerge consistently across implementations. Naming your Views after their purpose (not their number) helps your team immediately understand what each one does.

### Analyst View

Full dataset, complete pipeline, granular detail. This is the working View where transformation logic is built, tested, and validated. It typically contains the most tasks and the least filtering—it is built for depth, not presentation.

### Manager / summary View

Aggregated to the metrics that matter for operational decisions. Group & Aggregate tasks reduce rows to summary level. KPIs, period-over-period comparisons, and simplified column sets replace the transaction-level detail of the analyst View.

### Dashboard View

Pre-shaped for visualization. Columns are renamed to display-ready labels, date ranges are filtered early in the pipeline, and data is structured to match exactly what the chart or dashboard expects. Pipeline execution is kept lean so the dashboard loads fast.

### Export View

Schema-matched to a specific destination system. Data types are converted and validated, formatting meets the target system's requirements, and the View is typically connected to automated delivery. What goes out the door is exactly what the destination expects—no manual reformatting.

![Dataset open in the editor with three View tabs along the bottom and a data grid above](https://docs.mammoth.io/api/v1/images/20260409_082040_Onedatasetwiththreeviewtabs.jpg)

## Work with Views

### Create a View

Open a dataset from the Data Library. Click the **+** icon in the View tab bar at the bottom of the data editor. The new View opens immediately, starting with the full combined dataset and an empty pipeline.

Name the View before building the pipeline. A clear name set at creation prevents the accumulation of "View 3", "New View", "Copy of View 1" that makes datasets hard to navigate for the whole team.

![Data editor with View 1, View 2, and View 3 tabs and a Create new view tooltip on the plus icon](https://docs.mammoth.io/api/v1/images/20260326_102511_iconintheviewtabbar.jpg)

### Name Views

View names are the primary navigation tool for your team. Treat them as labels for a shared workspace, not personal shorthand.

Recommended pattern: **[Audience or Purpose]—[Destination or Frequency]**

![Dataset with four named View tabs: University USA, Sales, Marketing, and Q1-2025](https://docs.mammoth.io/api/v1/images/20260326_102527_Fournamedviewtabs.jpg)

| Good | Why |
| --- | --- |
| Finance Team—Monthly Close | Clear owner, clear cadence |
| Executive Dashboard—Daily | Clear audience, clear delivery |
| Salesforce Export—Automated | Clear destination, clear mode |
| TEST—New Attribution Logic | Clearly non-production |

| Avoid | Why |
| --- | --- |
| View 2 | No context |
| Analysis | Too generic |
| Copy of View 1 | Suggests unfinished work |
| temp | Temporary Views rarely get deleted |

Prefix test and staging Views consistently—**TEST—** or **STAGING—**—so team members immediately know not to rely on them for production outputs.

### Duplicate a View

Duplication creates an exact copy of an existing View: same pipeline tasks, same configuration, same settings. The copy is fully independent—changes to it never affect the original.

#### When to duplicate

- Testing new pipeline logic without touching a production View
- Creating a manager-level summary starting from an analyst View
- Building a variant formatted for a different destination
- Isolating a specific time period by duplicating and changing date filters

#### How to duplicate

Use the **More options** button on the View tab, or the three-dot menu on the View row in the dataset overview, and select **Duplicate**. The copy appears next to the original View.

![View Settings menu listing Data Update, Rename, Duplicate, Reset, Apply template, Save as a template, Delete, and more](https://docs.mammoth.io/api/v1/images/20260409_082111_Right-clickcontextmenuonaview.jpg)

### Rename and delete Views

Rename any View via the three-dot menu → **Rename**.

**Delete** a View from the same three-dot menu. Deletion is permanent—the pipeline configuration is lost. Before deleting, confirm no team member or downstream destination relies on the View. For Views that are complete but may be referenced later, rename with an **ARCHIVE—** prefix rather than deleting immediately.

Note: Delete is disabled when only one View exists in a dataset. **A dataset must always contain at least one View**.

### Reset a View

**Reset** removes everything applied to the View and returns it to a clean state. This includes all pipeline transformation tasks, explore card filters, sorting applied to the data grid, column formats, display changes, and any other elements configured on the View.

Use the three-dot menu on the View row in the dataset overview → **Reset**. A confirmation dialog appears before the reset is applied. This action cannot be undone.

Use Reset when a View needs to be rebuilt from scratch—for example, when the original pipeline logic is no longer valid after a source schema change, or when repurposing an existing View for a completely different use case. If you only want to remove presentation-layer changes such as column renames and formatting while keeping the pipeline intact, use the **Display Changes** panel at the bottom of the data editor instead.

### Save and apply View templates

Templates capture a View's pipeline as a reusable set of transformation rules, without the underlying data. Once saved, a template can be applied to any View to replicate the same pipeline—against the same schema or a different one.

**Save as a template.** Open the three-dot menu on the View row and select **Save as a template**. The template is downloaded to your system as a **.mammoth** file. This file contains the full pipeline structure and can be shared, stored, and reused across datasets and projects.

**Apply template.** Open the three-dot menu on the View row and select **Apply template**. This option is only active when the View's pipeline is empty—it is disabled if the View already has tasks. Selecting a template loads the saved pipeline rules into the current View.

**Schema matching and errors.** Apply template works best when the target dataset has the same column structure as the dataset the template was built on. If the schemas differ, the template still applies, but tasks that reference columns not present in the current dataset will show errors in the pipeline—for example, "This column is missing: Department." Each errored task presents a **Fix** or **Delete** option. You can work through these one by one to resolve mismatches or remove tasks that are not relevant to the new dataset.

#### When to use templates

- Applying the same transformation logic to multiple datasets from similar source systems
- Onboarding a new dataset that follows the same structure as an existing one
- Distributing a proven pipeline to other team members or projects as a file

## View Settings and Auto-sync

Access View Settings via the three-dot menu on any View row in the dataset overview.

### View Settings options

- **Open**—Open the View in the data editor (hidden when you are already in that View)
- **Settings**—Open the View's settings in the preview panel
- **View activity**—Show the View's activity
- **Auto-sync**—Turn the View's own Auto-sync on or off; the menu item shows the current ON or OFF state
- **Rename**—Edit the View display name
- **Duplicate**—Create an independent copy of the View and its pipeline
- **Reset**—Remove all transformations, elements, and filters, returning the View to a clean state
- **Apply template**—Apply a saved pipeline template to this View
- **Save as a template**—Save this View's pipeline as a reusable template
- **Delete**—Permanently remove the View (disabled when only one View exists)

The menu can include other items depending on your workspace. **Rename** is disabled when the View's pipeline has an error.

![View Settings menu opened from a View tab with Rename, Duplicate, Reset, template, and Delete options](https://docs.mammoth.io/api/v1/images/20260326_102611_ViewSettingsdropdown.jpg)

### How Auto-sync works

**Auto-sync** is a setting on each link between a View and the data it receives or sends. It controls whether data passes along that link automatically.

- **Auto-sync ON**: New data passes along the link and the View reruns without action.
- **Auto-sync OFF**: Data stops at that link. On a View's incoming link, the View is not rerun when its dataset receives a new batch, and the new batch stays pending until you release it.

The Auto-sync controls may not be available in every workspace.

#### Auto-sync options

To manage every link of a View in one place, open the View in the data editor, select **More Pipeline options** in the pipeline toolbar, then select **Auto-sync options**. The dialog has two tabs:

- **Incoming data**: the sources the View receives data from.
- **Outgoing data**: the destinations the View sends data to.

![Auto-sync options dialog with the Incoming data tab selected, listing the source dataset with its type, last sync time, and Auto-sync status](https://docs.mammoth.io/api/v1/images/20261009_013938_Auto-sync-options-modal.jpg)

Each row shows the source or destination, its type, the last sync, and the Auto-sync state (ON or OFF). Tick rows and select **Enable auto-sync** or **Disable auto-sync**, then select **Apply changes**.

You can also turn a single View's Auto-sync on or off from its menu (**Auto-sync**), or tick several Views in the dataset's Views list and use the **Enable auto-sync** and **Disable auto-sync** buttons.

### View status indicators

The dataset's Views list has an **Auto-sync** column for each View, and highlights if a View has encountered an error or has pending data.

![Data Library list with a View Settings menu showing Open and Settings items](https://docs.mammoth.io/api/v1/images/20260326_102631_Datasetoverview.jpg)

| Indicator | Meaning |
| --- | --- |
| **ON** | Data updates run automatically for the View. |
| **OFF** | New data is held as pending until you release it. |
| **Error icon (!)** | The View encountered an error during its latest execution. The icon appears next to the View name. |
| **Warning triangle (⚠)** | "Data may be inconsistent. Check Dataflow Monitor to manage Pending data updates." Select it to open the data status dialog. |

### When Auto-sync is off

When a View's incoming Auto-sync is OFF and its dataset receives a new batch, the View does not rerun and the batch is held as pending in **Monitor**, on the **Needs Attention** tab. The item reads "new batch received, auto-sync off".

To release pending data manually:

- **From Monitor**: Monitor → **Needs Attention** tab → select **Process** on the item (use **Preview** to look at the batch first). For a pending update to a destination, the main button is **Update**.
- **From the pipeline panel**: Three-dot menu on the Send to Dataset task card → **Run now**

Monitor also has a **Paused** tab that lists items paused through Auto-sync controls. Select **Turn Auto-sync ON** there to resume them.

#### When to use Auto-sync OFF

For Views feeding compliance or executive reports where output should be reviewed before delivery. For Views connected to production systems where an unverified pipeline change should not propagate automatically. For any scenario where a human needs to approve data before it reaches its destination.

For full documentation on Auto-sync and pending states, see [Monitor](https://docs.mammoth.io/learn/monitor/).

## View management best practices

**Keep each View to one purpose.** A View that serves three audiences with a 40-task pipeline is harder to maintain, harder to explain, and harder to modify safely than three focused Views. When a View becomes difficult to describe in one sentence, it is time to split it.

**Name first, build second.** The few seconds spent naming a View at creation saves minutes of confusion per team member per day as the dataset grows.

**Never modify a production View to test new logic.** Duplicate it, prefix the copy with TEST, experiment, validate, then either promote the result or discard it. Keeping production Views untouched during iteration helps keep pipelines reliable.

**Agree on naming conventions early.** Once a dataset has ten Views with inconsistent names, the cost of reconstruction—figuring out what each View does by reading its pipeline—is significant. Establish the convention before it is needed.

**Save templates from proven Views, not works in progress.** A template applied across multiple datasets carries any errors in its pipeline to each new application. Validate the source View thoroughly before saving it as a template.

## Troubleshoot Views

### Destination not updating despite pipeline running

Check Auto-sync status in View Settings. If the crossed-out sync icon is visible, Auto-sync is OFF and the pending output is held there. Open [Monitor](https://docs.mammoth.io/learn/monitor/) → **Needs Attention** and select **Process** (or **Update** for a pending destination update) to release, or toggle Auto-sync ON to resume automatic delivery.

### View not reflecting the latest batch data

Check the Last Updated timestamp in the dataset overview. If it predates the batch change, the pipeline has not re-run. Open the View and click **Update Data** on the relevant pipeline step to run it manually. If the timestamp is current but data is missing, check the pipeline for Conditional Filter tasks that may be excluding the new batch's rows. For batch-level troubleshooting, see [Batch Management](https://docs.mammoth.io/learn/data-sources/batch-management).

### Pipeline error after a batch was added or changed

The error icon next to the View name indicates the last pipeline run failed. Open the View—the failing task is highlighted in the pipeline panel with error detail. Common causes after a batch change are column name mismatches or data type conflicts between the new batch and what the pipeline expects. For schema consistency guidance, see [Batch Management](https://docs.mammoth.io/learn/data-sources/batch-management).

### View performance is slow

Apply a Conditional Filter as the first pipeline task to reduce data volume before downstream processing. Enable Draft mode before adding multiple tasks to avoid repeated full pipeline executions during development. For detailed pipeline optimization guidance, see [Pipeline Optimization](https://docs.mammoth.io/learn/performance/pipeline-optimization).

### Team member cannot find a View

Views are visible to all project members (Project Admins and Project Editors). If a View appears missing, confirm the team member has the correct project access. Views cannot be made private within a project.

### Apply template option is greyed out

Apply template is only enabled when the current View's pipeline is empty—it's disabled as soon as the View has any pipeline tasks. To use it, apply the template to a fresh View, or Reset the current View first if you want to replace its pipeline. (Templates themselves aren't scoped to a project or organization—they're `.mammoth` files saved to your computer via **Save as a template**, so any template file you have access to can be applied.)

---
Source: https://docs.mammoth.io/learn/data-sources/view-management.md · Updated: 2026-10-13