You identified nodes in the Northwind dataset. The Import tool provides a visual interface for defining your data model and loading data without writing code.
In this lesson, you will learn how the Import tool is organized, the workflow for importing data, and how mapping works.
This workshop uses the Import tool
The Import tool is ideal for this workshop:
Visual interface - See your model as you build it
Tabular files - Northwind data is in CSV format
Small dataset - Under 1M rows (91 customers, 830 orders, 77 products)
No code required - Focus on concepts, not syntax
Other import methods
Neo4j offers other import methods for larger datasets or different sources. See the Data Import documentation for options like LOAD CSV, neo4j-admin import, and ETL tools.
Understanding the import tool
The Import tool in Neo4j Aura helps you load data from various sources into your graph database without writing any code.
Cloud storage (AWS S3, Azure Blobs & Data Lake Storage, Google Cloud Storage)
Local files (CSV)
Key benefits:
Visual - Design your graph model with a drag-and-drop interface
No coding required - Map tables and columns to graph elements without writing Cypher
Validated - Preview your data before importing
Fast - Optimized for bulk data loading
You’ll find the Import tool in the left navigation of your Aura instance.
The import tool
The Import tool consists of three main tabs that reflect the three stages of importing.
Exploring the Import Tool UI
The Import tool has three main tabs:
Data Sources tab - Where you upload files or connect to databases
Graph Models tab - Where you design your graph model on the canvas
Import Jobs tab - Where you view import history and results
1. Data Sources tab
The Data Sources tab is where you connect to your data.
Upload local CSV/TSV files
Connect to databases or cloud storage
Preview columns and sample data
Manage multiple data sources
2. Graph Models tab
The Graph Models tab is where you design your graph.
Visual canvas for creating nodes and relationships
Map data sources to graph elements
Configure properties and IDs
Preview your model before importing
3. Import Jobs tab
The Import Jobs tab shows your import history.
View import progress and results
Review nodes created and properties set
Check for errors or warnings
Re-run previous imports
Connecting to data sources
To choose a data source, navigate to the Data sources tab and click New data source.
From there, you can choose the source of your data.
Depending on which data source you choose, you will need to configure the data source and provide the credentials to connect to the data source.
The data source you choose must be publicly accessible.
For CSV and TSV files, you will be prompted to upload the files.
Using the graph models tab
Once you have a data source connected, switch to the Graph Models tab to design your model.
The Canvas Layout
The canvas has three panels:
Left panel - Data sources (your uploaded files)
Center panel - Model canvas (where you build your graph)
Right panel - Details panel (configure selected nodes/relationships)
The Canvas: Data Source Panel (left)
The Data Source Panel shows your connected data.
View all uploaded files or connected tables
Expand each source to preview columns
See sample data values
The Canvas: Model Canvas (center)
The Model Canvas is your visual workspace.
Add nodes using the + Node Label button
Create relationships by dragging between nodes
Green checkmarks indicate complete mappings
The Canvas: Details Panel (right)
The Details Panel configures selected elements.
Set the node label or relationship type
Map to a data source table
Configure properties and data types
Set the unique ID property
Creating and labeling nodes
How to create a node:
Click the + Node Label button (bottom center of the data model pane)
A node appears on the canvas
Type the label directly on the node, or use the Label field in the details panel
To complete the mapping process, every node needs a label, a data source, at least one property, and an ID property.
The green checkmark against each node and relationship indicates that all requirements are met and the data is ready import.
Mapping tables to nodes
To map a node to a data source:
Select your node on the canvas
In the details panel, find the Table dropdown under the Definition pane
Select the data source to map to this node
Choose Map properties from table to auto-populate properties
Configuring properties
Once a data source is mapped, you’ll see properties listed in the Definition pane of the details panel.
For each property, you can:
Rename - Click the pencil icon to change the property name
Change type - Select the data type from the dropdown
Select Column - select the column from the data source to map to this property
Available data types
The Import tool auto-detects data types from a sample of your source data. You can change the data type of a property from the dropdown.
String - Text values
Integer - Whole numbers
Float - Decimal numbers
Boolean - true/false values
Date - Date values (e.g., order dates)
DateTime - Date and time values (e.g., timestamps)
Neo4j also supports more complex types like lists and maps. For deeply nested data (like lists of lists or maps of maps), start by asking what questions you need to answer. Depending on your use case, nested values might become separate nodes, stay as properties, or be stored externally with a pointer in the graph.
Setting the ID property
The ID property uniquely identifies each node in the source data.
To set the ID property, select the column from the list.
Every node must have a unique identifier. This ensures that each node is unique and can be identified within the source data.
Automatic ID selection
The Import tool attempts to automatically select columns with "id" in the name, but you should verify this is correct for your data.
Creating relationships
To create a relationship between two existing nodes:
Create and map both nodes you want to connect
Select one of the nodes on the canvas (the from node)
Hover over the selected node - a grey circle with a green plus sign appears (the halo)
Drag the halo to the other node and release
A relationship line is drawn between the two nodes
Once the relationship is created, you can configure its properties in the details panel.
Create a node and relationship together
You can also create a relationship to a new node in one step by dragging the halo to an empty space on the canvas.
Configuring relationship properties
Configure the relationship in the Definition pane of the details panel.
Required relationship settings:
Type - Name the relationship type (e.g., PURCHASED, IN_CATEGORY)
Table - Select the data source that contains the relationship data
Node ID mapping - Specify which nodes are connected
From - Identifies the source node using a column that matches the source node’s ID property
To - Identifies the target node using a column that matches the target node’s ID property
Properties - Use the Map from table to set relationship properties
Relationship direction
All relationships have a direction in Neo4j. If you need to reverse the direction, select the relationship and use the reverse direction button in the details panel.
Previewing the import
To check your model before importing, you can preview your data using the Preview icon.
Click the Preview icon
Choose Preview all or Preview selected
The preview shows a sample of your data as a graph
Running the import
When you are ready to import your data, click the Run Import button.
Once completed, you’ll see a results summary showing nodes created, properties set, time elapsed, and data processed.
For longer running imports, you can view the progress by navigating to Import and clicking the Import Jobs tab.
Viewing import results
Once the import is complete, a summary of the import will be displayed.
You can expand each element in the results to see the generated Cypher. This provides insight into how the Import tool constructs statements.
Following best practices
Before importing:
Review your source data for quality and completeness
Identify unique identifiers in each data source
Plan your node labels and property names
Understand the graph structure you’re building
During modeling:
Use descriptive, singular labels (Product not products)
Follow naming conventions (CamelCase for labels, camelCase for properties)
Set appropriate data types for properties
Always verify the ID property is correct
After import:
Verify the data with Cypher queries
Check node counts match expected values
Review the results summary for any warnings
Test sample queries to ensure data is correct
Try for yourself
Now it is time to try for yourself. Head to the next lesson to import your first nodes with step-by-step guidance.
Summary
In this lesson, you learned how the Import tool works:
Three-tab structure - Data Sources, Graph Models, and Import Jobs
Data source options - Databases, data warehouses, cloud storage, and local files
Three-stage workflow - Provide data → Model and map → Run import
Graph modeling - Create nodes on the canvas and set labels
Mapping process - Connect tables/files to nodes, regardless of source type
Property configuration - Select columns, rename properties, and set data types
ID property - Every node must have a unique ID property for node identification
Creating relationships - Use the halo to connect existing nodes or create new nodes with relationships
Relationship configuration - Set type, map data source, configure Node ID mappings (From/To), and select properties
Preview functionality - Test your model with sample data before importing
Import execution - Batch processing with progress tracking and results summary
In the next lesson, you will use the Import tool to import your first nodes with step-by-step guidance.
