> ## Documentation Index > Fetch the complete documentation index at: https://support.entegrata.com/llms.txt > Use this file to discover all available pages before exploring further. # Automap a Pipeline > Step through every unmapped silver data type in a pipeline with Automap, in dependency order ## Overview Running **Automap** from a pipeline steps you through each unmapped silver data type in that pipeline, one at a time. Data types come in dependency order, so a data type is mapped before any data type that depends on it. On each data type, Automap picks or suggests a primary source and loads field suggestions, which you then review in the familiar [mapping editor](https://support.entegrata.com/data/pipeline/automap-entity.md#review-suggestions). To learn how suggestions are rated, see the [Automap Overview](https://support.entegrata.com/data/pipeline/automap-overview.md). > [!NOTE] > > **Required permissions** > > * **Data Pipelines**: Access > * **Manage Production Pipelines** or **Manage Dev Pipelines**, for the pipeline: Update > * **Manage Silver Entities**: Update > > See [roles and permissions](https://support.entegrata.com/administration/user/roles-and-permissions.md#feature-reference). ## Start a Session **1. Open the pipeline list** Go to **Pipelines** to see your pipelines. **2. Choose Automap** Open the options menu (**⋮**) on the pipeline's row and select **Automap**. The session opens on the first data type that's 10% mapped or less. Data types that are more than 10% mapped are left out. The breadcrumb reads **Mapping / \ / Automap**. A stepper at the top shows your progress through the session. For the current data type it shows its mapping status and which data types it **depends on** and is **needed for**. ![Automap session on a data type, with the stepper, status line, skipped count, and Skip and Next buttons](https://mintcdn.com/entegrata/TpvgrAx_WL1xNLct/images/pipelines/automap_pipeline_session.png?fit=max\&auto=format\&n=TpvgrAx_WL1xNLct\&q=85\&s=aebc6aa084750777339fc46afb241cfa) ## Map Each Data Type **1. Let Automap find a primary source** If the data type has no primary source, Automap looks for one. * If one candidate is a clear match, Automap selects it and saves it right away. * Otherwise, **Select Primary Data Source** opens with candidate tables ranked by confidence. Choose the table that holds one row per record and click **Confirm Selection**, click **Browse for Source** to search the catalog, or click **Skip** to move on. ![Select Primary Data Source, with candidate tables ranked by confidence](https://mintcdn.com/entegrata/TpvgrAx_WL1xNLct/images/pipelines/automap_select_primary_source.png?fit=max\&auto=format\&n=TpvgrAx_WL1xNLct\&q=85\&s=3e9b82f6ac67cf5ed626f90764c92a14) If Automap can't tell which columns identify a record, **Configure Primary Source** opens. Under **Primary Key Fields**, select one or more fields that uniquely identify each record, then click **Confirm**. **2. Review field suggestions** The status line reads **Loading field suggestions…**, then **Review suggestions, then save or skip**. Accept, dismiss, and configure joins as described in [Automap in the Mapping Editor](https://support.entegrata.com/data/pipeline/automap-entity.md#review-suggestions). **3. Save** Click **Save**. **Next** stays disabled until you save or skip the current data type. **4. Go to the next data type** Click **Next**. Automap loads suggestions ahead of time for the next few data types, so they open quickly. To go straight to a particular data type, open **Jump to entity** and filter the list by name. ## Skip a Data Type Click **Skip** to move past a data type without mapping it. Skipping doesn't save its field suggestions, but a primary source Automap already saved for it stays saved. **Skip this entity?** asks how long to skip it: * **For this session**: leave the checkbox clear. The data type returns the next time you start a session. * **For future sessions**: select **Don't surface this entity again in future auto-mapping sessions**. The data type stays skipped until it's mapped or you restore it. This setting is saved in your browser, so it doesn't follow you to another browser or device. To bring back a skipped data type, click the **skipped** count in the toolbar, then click **Restore** next to it. ## Finish After the last data type, the completion screen shows **All done.** with the number of data types mapped and skipped. Click **Back to Pipeline** to return to the pipeline. Mappings you saved during the session are ordinary saved mappings. [Publish](https://support.entegrata.com/data/pipeline/publishing-mappings.md) and [deploy](https://support.entegrata.com/data/pipeline/deploying-pipelines.md) them as usual. If you refresh the page mid-session, your place, your skips, and your completed data types are kept. Starting a new session from the pipeline list resets them, except for data types you skipped for future sessions. ## Troubleshooting | Problem | What to do | | - | - | | **Pipeline is fully mapped** | Every data type is more than 10% mapped. Use [Automap in the Mapping Editor](https://support.entegrata.com/data/pipeline/automap-entity.md) on individual data types instead. | | **Nothing to step through** | The pipeline has no data types yet. [Add one](https://support.entegrata.com/data/pipeline/managing-entities.md) first. | | **Dependency cycle detected** or **Can't determine step-through order** | Some data types depend on each other in a cycle. Resolve the cycle, then start again. | | **Entity no longer in pipeline** | The data type was deleted or removed from the pipeline after the session started. Click **Go to next entity** or **Exit**. | | **Nothing to map — click Skip to continue** | Automap found no suggestions for this data type. Skip it, or map it by hand later. | | **Automap failed** | Click **Retry**. If it keeps failing, contact [support](https://support.entegrata.com/index.md). | | **Automap** isn't in the pipeline's options menu | Your role needs Update on this pipeline. See [Overview](#overview) for the required permissions. | ## Related Topics > [card] [Automap Overview](https://support.entegrata.com/data/pipeline/automap-overview.md) > > How suggestions are rated and applied > [card] [Automap in the Mapping Editor](https://support.entegrata.com/data/pipeline/automap-entity.md) > > Run Automap on a single data type > [card] [Managing Data Types](https://support.entegrata.com/data/pipeline/managing-entities.md) > > Add and organize data types in a pipeline > [card] [Publishing Mappings](https://support.entegrata.com/data/pipeline/publishing-mappings.md) > > Publish saved mappings ## Related topics - [Automap Overview](https://support.entegrata.com/data/pipeline/automap-overview.md) - [Automap in the Mapping Editor](https://support.entegrata.com/data/pipeline/automap-entity.md) This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.