If your map has a layer with a large number of points, you can aggregate your data using clustering to extract meaningful information from the data. When you enable clustering, Map Viewer groups point features that are within a certain distance of one another on the map into one symbol. Clustering allows you to see patterns in the data that are difficult to visualize when a layer contains hundreds or thousands of points that overlap and cover each other.
If your layer is styled by category, clusters show the predominant category in each cluster. You can also choose to use pie chart clusters to show the proportions of categories in each cluster.
Tip:
For recommendations on applying clustering to high-density datasets, see Best practices for visualizing high-density data.
Clusters are represented by proportionally sized symbols based on the number of point features in each cluster. Smaller cluster symbols have fewer points, while larger cluster symbols have more points. You can adjust the size range applied to the cluster symbols.
Clustering is applied dynamically at multiple scales, which means that as you zoom out, more points are aggregated into fewer groups, and as you zoom in, more cluster groups are created. When you zoom to a level at which the clustering area around one point feature no longer contains other features, that point feature is not clustered; it is displayed as a single point feature with the styling options applied to the layer. You can adjust the number of point features grouped into clusters by setting the cluster radius.
Nonchart clusters appear on the map with a default label. All types of clusters have a default pop-up that appears when you click a cluster on the map. You can customize the cluster pop-up and labels.
Enable clustering
When you enable clustering on a point layer, clusters automatically appear on the map using a default configuration. As you make changes to the cluster settings, the changes are immediately shown on the map. This allows you to experiment with cluster radius and other settings to achieve the desired look and message of your map.
To enable clustering on a point layer, do the following:
- Confirm that you are signed in and, if you want to save your changes, that you have privileges to create content.
- In Map Viewer, open the map containing the layer or add the layer directly.
- On the Contents (dark) toolbar, click Layers
. - In the Layers pane, select the layer on which you want to enable clustering.
- On the Settings (light) toolbar, click Aggregation
. - In the Aggregation pane, turn on the Enable aggregation toggle button.
- Choose one of the following:
- Clustering—Aggregates points into clusters and displays them as one symbol.
- Clustering (chart)—Aggregates categorized points into pie chart cluster symbols.
Note:
Chart clustering is only available if the layer is styled by Types (unique symbols), Types and Size, or Counts and amounts (color) using classification.
- On your chosen clustering style, click Options and do any of the following to change the appearance of the clusters:
For Clustering, optionally turn on the Override cluster symbol toggle button.
Clusters inherit the style options you configure for the layer. Turning on the Override cluster symbol toggle button allows you to style the clusters separately. Overriding the cluster symbol can help differentiate clusters from unclustered features in the same layer. See Use style options for more information on changing the symbol style.
- For Clustering (chart), adjust the Shape slider to change the chart appearance from pie to donut.
- Adjust the Cluster radius slider to change the number of features that are grouped into a cluster. Specify a lower cluster radius to group fewer features into each cluster. Specify a higher cluster radius to group more features into each cluster.
- Adjust the Size range slider to set the minimum and maximum cluster size.
- Add and edit fields to use in the clusters.
- Customize the cluster labels.
- Customize the cluster pop-ups.
- Close the Aggregation pane when you're finished.
Configure cluster fields
When you enable clustering, the clusters are created using one or more default fields. You can edit these fields, add new summary statistic fields for use in labels and pop-ups, and change the formatting. For example, you can change the number of decimal places that appear for a numeric field.
To configure cluster fields, do the following:
- Follow the steps of the Enable clustering section above.
- In the Clustering pane, click Cluster fields.
- Optionally, expand a default field and specify its properties:
- Type a field alias in the text box.
- If the default field is numeric, under Formatting, turn the Show thousands separator toggle button on or off.
- To add a new field, do the following:
- Click Add field.
The field configuration options appear.
- Choose a data field from the drop-down menu.
- Choose a statistic type from the drop-down menu.
- Optionally, type a field alias in the text box.
- Under Formatting, choose the number of decimal places from the drop-down menu and turn the Show thousands separator toggle button on or off.
- Click Add field.
- To delete a field, click Options
and choose Delete.Note:
The default fields used to create the cluster—for example, count and predominant value for a pie chart cluster—cannot be deleted.
The fields you configure are available for use in labels and pop-ups. Formatting changes are automatically applied to fields that are currently in use for the clusters.
Configure cluster labels
Labeling clusters is similar to labeling individual features in a layer. You control the label style—font, text size, placement, and so on. You can keep the labels simple by showing the number of features in each cluster, or, if the layer is styled using an attribute, you can use this attribute for the cluster label. For example, if the layer shows parcels by their value per square foot, you can configure the cluster label to show the average value per square foot of all the points in each cluster. Map Viewer also allows you to use a label filter to only display labels on clusters that meet the filter criteria.
Configuring label classes allows you to further customize cluster labels. For example, you can use label classes to label each cluster based on two attributes, such as the average earthquake magnitude and the earthquake count, using a different label style for each attribute. Cluster label classes are configured in the same way as label classes for features.
Note:
Any unclustered point feature displays a feature label if feature labels are enabled for the layer.
To configure cluster labels, do the following:
- Follow the steps of the Enable clustering section above.
- In the Clustering pane, click Cluster label.
- In the Label features pane, turn on the Enable labels toggle button.
- Click Add label class to configure label classes, and specify the options for each class:
Note:
See Configure labels for more information about label configuration.
- Click the Label field selector, select a different field for the label, and click Replace.
The default label field is the number of features in the cluster. If the layer is styled using an attribute, you can use the attribute as a label. If you want to use a different summary statistic as a label, add a cluster field.
Tip:
You can use a custom attribute expression written in Arcade instead of an attribute field. Click Use expression
and use the editor window to create the expression. You can also use existing expressions to build new expressions; however, some variables may not work in all profiles—for example, an expression created for pop-ups may not work for styles. To use an existing expression, select it from the Suggestions
tab under Existing. - Click Edit label filter, click Add expression, and set up a filter fo display labels on a subset of clusters. For more information, see Apply filters.
- Click Edit label style. In the Label style window, change the label style options, including the font, text size and color, placement, offsets, and halo effect.
- Adjust the Visible range slider to change the visible range for the cluster labels.
- Click the Label field selector, select a different field for the label, and click Replace.
Configure cluster pop-ups
Cluster pop-ups appear when you click a cluster on the map. The information displayed in cluster pop-ups depends on the style applied to the layer. For example, if the Predominant category style is applied, the default cluster pop-up includes the value of the predominant attribute for each cluster. You can customize the cluster pop-ups to suit your needs.
Note:
Clicking an unclustered point feature displays a feature pop-up if feature pop-ups are enabled for the layer.
To configure cluster pop-ups, do the following:
- Follow the steps of the Enable clustering section above.
- In the Clustering pane, click Cluster pop-ups.
- In the Pop-ups pane, customize the cluster pop-ups by doing any of the following:
Note:
See Configure pop-ups for more information about pop-up configuration.
- Optionally, click Attribute Expressions to create and modify Arcade expressions to use in the cluster pop-up.
- Expand the Title section and type a title for the pop-up. You can enter static text and click Add field
to select attribute fields from the data. - Click the Text content block and click the cluster pop-up text. Change the default content and click OK.
- Click Add content, click the content type, and add content and media to the cluster pop-up, including images and attribute fields.
- Drag the content blocks to new positions in the pop-up to change the order.
- Click the clusters on the map to view the pop-ups.
To learn more about the individual features in a cluster, click Browse features on the pop-up and click the forward and back arrows to scroll through information for each feature. To go back to the summary information for the cluster, click the record count between the arrows.
Tip:
You can click Dock
on the pop-up to lock the pop-up display to the top of the map. Click Undock 
to display pop-ups near their associated clusters.
Considerations
Keep the following in mind:
- Clustering is supported for hosted point feature layers, ArcGIS Server 10.3.1 and later point feature layers that support pagination, CSV layers, and feature collections.
- Clustering is not supported for multipoint feature layers.
- The Heat Map style is not supported for layers with clustering enabled.
- Publishing hosted tile layers from hosted feature layers is not supported for layers with clustering enabled.
- Clustering is unavailable during an edit session and when using feature effects.
- Override cluster symbol is not available for the Clustering (chart) style. Unclustered features are symbolized as charts.
- Clustering (chart) supports a maximum of 10 slices per chart. If there are more than 10 categories in a cluster, the top 9 categories appear in the chart and all other values are grouped into Other.