Data table fields define how information is stored and displayed in Atlas. Each field type serves specific data needs and can be enhanced with AI capabilities.
Field Types
Text Fields
| Field Type | Description |
|---|---|
| Text | Store names, descriptions, and general text |
| URL | Web links with click-to-open functionality |
| Image | Image URLs with preview display |
| PDF document URLs with preview support |
Numeric Fields
| Field Type | Description |
|---|---|
| Number | Integer and decimal values |
| Currency | Monetary values with currency formatting |
| Percentage | Percentage values with % display |
| Rating | Numerical evaluation scales (e.g., 1-5 stars) |
Date & Time Fields
| Field Type | Description |
|---|---|
| Date | Calendar dates with timezone support |
| Time | Time values |
| Duration | Time duration values |
Selection Fields
| Field Type | Description |
|---|---|
| Single select | Choose one option from a predefined list |
| Multi select | Choose multiple options from a predefined list |
Selection fields support custom colors for each option, enabling visual categorization on maps.
Boolean Fields
| Field Type | Description |
|---|---|
| Checkbox | True/false values for yes/no data |
Contact Fields
| Field Type | Description |
|---|---|
| Email addresses with validation and click-to-email | |
| Phone | Phone numbers with international format support |
User Fields
| Field Type | Description |
|---|---|
| User | Assign team members to records |
| Created by | Auto-populated with the user who created the record (read-only) |
| Modified by | Auto-populated with the last user to edit the record (read-only) |
Relation Fields
| Field Type | Description |
|---|---|
| Relation | Link records to another dataset |
| Lookup | Display values from related records (read-only) |
Relation fields connect datasets together, while Lookup fields retrieve values from linked records. See Relations & Lookups for the complete guide including setup, use cases, and best practices.
Formula Fields
| Field Type | Description |
|---|---|
| Formula | Computed values based on other fields (read-only) |
Formula fields calculate values dynamically using expressions, functions, and operators. See Formula Fields for the complete guide including syntax, functions reference, and examples.
System Fields
These fields are automatically calculated and read-only. The available spatial fields depend on your layer's geometry type:
| Field Type | Description | Geometry Types |
|---|---|---|
| Created at | Timestamp when the record was created | All |
| Modified at | Timestamp when the record was last modified | All |
| Area | Calculated area of the geometry | Polygons only |
| Perimeter | Calculated perimeter of the geometry | Polygons, Lines |
| Latitude | Computed latitude coordinate of the feature | Points only |
| Longitude | Computed longitude coordinate of the feature | Points only |
Atlas automatically shows only the spatial system fields relevant to your data — for example, Area and Perimeter won't appear on point layers, and Latitude/Longitude won't appear on polygon layers.
Geographic Fields
Geographic fields are typically auto-generated from geocoding and are read-only:
| Field Type | Description |
|---|---|
| Geometry | Raw geometry data (WKT/GeoJSON) |
| Address | Full street address |
| City | City name |
| State | State or province |
| Country | Country name |
| Country code | ISO country code |
| Region | Geographic region |
| Zip code | Postal/ZIP code |
| Latitude | Decimal latitude coordinate |
| Longitude | Decimal longitude coordinate |
| Coordinates | Combined X/Y coordinate pair |
| H3 index | Uber H3 hexagonal index |
| UTM zone | UTM coordinate zone |
| UTM easting | UTM easting value |
| UTM northing | UTM northing value |
| UTM hemisphere | UTM hemisphere (N/S) |
AI-Enhanced Fields
Enable AI for supported field types to automate data entry and analysis.
Supported Field Types
AI capabilities are available for:
- Text
- Number
- Date
- Checkbox
- Rating
- Phone
- Single select
- Multi select
Setup Process
- Add field → Select type → Toggle Use AI
- Write specific instructions with context
- Reference other fields using
{field_name} - Test with preview before creating
AI Capabilities by Field Type
| Field Type | AI Capabilities |
|---|---|
| Text | Generate descriptions, extract information, summarize |
| Number | Calculate values, estimate missing data, parse numbers |
| Checkbox | Make logical decisions, classify data |
| Single select | Auto-categorize, standardize inputs |
| Multi select | Tag and classify with multiple categories |
| Date | Extract dates from text, calculate relative dates |
| Rating | Score based on criteria, assess risk levels |
| Email/Phone | Extract and validate contact information |
Writing Effective AI Instructions
- Be specific about expected output format
- Provide context and boundaries
- Include examples of desired results
- Use the
{}button to insert field references
Field Configuration
Adding New Fields
- Open the data table view for your dataset
- Click + Add Field in the table header
- Select field type from the dropdown
- Configure field properties:
- Field name and description
- Validation rules and constraints
- Default values and formatting
- AI settings (if applicable)
- Click Create to save
If the field name you enter is already used by another column in the dataset, Atlas automatically appends a number to make it unique (e.g. "Status 2").
Modifying Existing Fields
Click the field header dropdown to access field settings:
- Rename field
- Change description
- Modify display formatting
- Add or remove AI capabilities
- Reorder fields (drag and drop)
- Delete field
Deleting Fields
When you delete a column, Atlas checks whether anything else depends on it — lookups, computed columns, style bindings, widget configurations, views, and form conditions. If dependencies are found, a deletion impact preview appears before the final confirmation. The modal groups impacts into:
- Will be deleted — dependent columns (lookups, computed fields) that reference this column and will be removed along with it.
- Will be cleaned up — references in layer styles, filter expressions, popups, and widget configs that will be automatically stripped.
After reviewing, confirm to proceed. The cleanup happens in a single transaction so the dataset is never left in a partially-broken state.
Read-Only Fields
The following field types cannot be manually edited:
- System fields (Created at, Modified at, Area, Perimeter)
- User tracking fields (Created by, Modified by)
- Lookup fields (values come from related records)
- Formula fields (values are computed)
- Geographic fields (generated from geocoding)