How do I visualize data in Neo4j Bloom?

Visualizing data in Neo4j Bloom turns your graph database into an interactive, visual story that’s easier to explore, explain, and share with others. Instead of reading raw Cypher or tables of nodes and relationships, you can see patterns, clusters, and paths emerge on a canvas in real time.

In this guide, you’ll learn how to visualize data in Neo4j Bloom step by step—from connecting to a database and running your first search, to styling nodes, building scenes, and sharing insights with your team.

1. Connect Neo4j Bloom to your database

Before you can visualize anything, Bloom needs to be connected to a running Neo4j database.

1.1 Choose a Neo4j database

You can use any of the following:

Neo4j Aura (cloud) – Fully managed, including free tiers.
Neo4j Sandbox – Temporary, pre-populated or blank instances:
- Go to https://sandbox.neo4j.com and create a project.
Self-hosted Neo4j – Local or server installation (Community or Enterprise).

Make sure the database is running and that you have:

Bolt URL (e.g., neo4j+s://... or bolt://localhost:7687)
Username (often neo4j)
Password

If you don’t have a long-lived database yet and want something quick for experimentation, start with:

Neo4j Sandbox: https://sandbox.neo4j.com
Or sign up for Neo4j Aura: https://console.neo4j.io

1.2 Launch Neo4j Bloom and connect

Open Neo4j Bloom (desktop app or browser version if available).
On the home view, choose New connection or Add data source.
Enter:
- Connection name (anything meaningful, e.g., “Product Graph”)
- URL (Bolt or Neo4j URI)
- Username and password
Click Connect.
Select the database if you have multiple (e.g., neo4j, graph, etc.).

Once connected, Bloom will be able to query your graph and display it visually.

2. Understand the Neo4j Bloom interface

Before visualizing data, it helps to know the main areas of the Bloom UI:

Search bar – The central input for phrase-based search and Cypher.
Perspective panel – Defines which node labels, relationship types, and properties are visible and how they appear.
Canvas – The graph visualization area where nodes and relationships are rendered.
Scene panel – Saved visual states (scenes) you can switch between or share.
Inspector / Details pane – Shows properties of selected nodes and relationships.

Visual exploration in Bloom revolves around three core actions:

Search (get data on the canvas)
Expand and refine (traverse relationships, filter, and group)
Style and save (change appearance, capture scenes, share perspectives)

3. Get your first visualization with search

The easiest way to start visualizing data in Neo4j Bloom is via the search bar.

3.1 Use phrase-based search

Bloom supports natural-language-like search patterns based on its Perspective. Examples of what you might type:

Person named "Alice"
Movies with title "The Matrix"
Orders in 2024
Product with id 123

When you run a search:

Bloom translates your phrase into a query based on the graph model defined in the Perspective.
The matching nodes appear on the canvas, visually laid out.
You can then click nodes to inspect their properties or expand connected data.

If nothing shows up, you may need to:

Adjust your Perspective to include the relevant labels and properties.
Simplify your phrase (e.g., search only for "Alice" or Person).

3.2 Use Cypher queries directly

For more control, you can enter Cypher in the search bar:

MATCH (m:Movie)-[:ACTED_IN]-(p:Person)
WHERE m.title = "The Matrix"
RETURN m, p

Any nodes and relationships returned by the query appear on the canvas.

Tips for visualization-ready Cypher:

Always RETURN the nodes and relationships you want to see.
Limit huge results, e.g., LIMIT 100, to keep the visualization readable.
Use WITH and filtering to focus on key patterns.

4. Explore and expand your graph on the canvas

Once you have nodes on the canvas, the real power of Bloom comes from interactive exploration.

4.1 Expand nodes to see connected data

To explore the neighborhood around a node:

Select one or more nodes on the canvas.
Right-click (or use the context menu) and choose Expand.
Bloom shows available relationship types (e.g., ACTED_IN, FOLLOWS, PURCHASED).
Select which relationships to expand to reveal connected nodes.

This lets you visually trace paths such as:

People connected to a movie.
Products in an order.
Accounts linked by transactions.

You can repeat the expand action to traverse multiple hops and uncover larger structures.

4.2 Collapse or hide nodes for clarity

As you explore, the canvas can become cluttered. To keep the visualization useful:

Hide nodes/relationships that aren’t relevant:
- Select them, right-click, and choose Hide.
Collapse expanded branches:
- Select a node and choose Collapse to fold its neighborhood back.

This helps you maintain a focused visual narrative around the part of the graph that matters now.

4.3 Select and inspect nodes

Click on a node or relationship to:

View properties (id, name, timestamps, categories, etc.)
See labels and relationship types
Use context actions (expand further, filter, style)

You can select multiple elements using:

Shift-click (expand selection)
Drag-select (lasso selection on the canvas)

5. Use filtering and grouping to refine visuals

Bloom provides powerful ways to refine what you see, which is crucial for understanding complex graphs.

5.1 Filter by labels, relationships, or properties

You can filter the current scene to focus on specific subsets:

Filter by node label (e.g., show only :Customer and :Product).
Filter by relationship type (e.g., only :PURCHASED edges).
Filter by properties, such as:
- age > 30
- status = "active"
- amount > 1000

Filtering reduces noise and highlights patterns such as:

High-value customers.
Suspicious transactions.
Communities around a specific topic or entity.

5.2 Group nodes by categories

Grouping lets you visually cluster nodes that share something in common:

Group by label (all :Order nodes grouped together).
Group by a property (e.g., country, segment, status).
Group by relationship pattern (e.g., all orders linked to the same customer).

Bloom can represent groups as:

Super-nodes (collapsed representations of many nodes).
Clustered layouts that keep related items visually close.

This makes patterns like regional clusters or category-based communities easy to spot at a glance.

6. Style your visualization for clarity and impact

A key part of visualizing data in Neo4j Bloom is configuring how nodes and relationships look. Styling helps different entities stand out and makes the graph more readable.

6.1 Style through Perspectives

Visual appearance in Bloom is primarily controlled through Perspectives:

A Perspective defines:
- Which labels and relationships are visible.
- The icon, color, and size of each node type.
- How properties are displayed in search and on the canvas.

For example, you might configure:

:Person nodes as blue circles with the person’s name as the caption.
:Company nodes as green hexagons with companyName as the caption.
:TRANSACTION relationships as red arrows with amount displayed.

To adjust these:

Open the Perspective editor (depending on your Bloom version, via the side panel or menu).
Choose a node label (e.g., Person).
Set:
- Caption property (what text is shown on the node).
- Icon and color.
- Optional size mapping based on a numeric property (e.g., more important nodes larger).

6.2 Apply layouts

Bloom supports different graph layouts, such as:

Force-directed (natural “network” layout)
Radial (nodes arranged around a center)
Hierarchical or layered layouts (if supported in your version)

Try different layouts to:

Reveal community structures.
Show hierarchical relationships.
Center the graph around a key node (e.g., a customer or account).

You can usually access layout options via a toolbar or context menu; apply them to all nodes or a selected subset.

6.3 Highlight patterns and paths

For storytelling and analysis, it’s helpful to highlight:

Shortest paths between nodes.
Key hubs (high-degree or central nodes).
Specific relationship types.

You can do this by:

Selecting a subset of nodes and applying different colors or sizes.
Using Bloom’s “path” or “expand to path” features where available.
Saving styled views as Scenes (covered next).

7. Save and share your visual insights with Scenes

To capture and reuse your visualizations, Bloom uses Scenes, which are snapshots of your current canvas state.

7.1 Create a scene

Once you’ve arranged and styled a view:

Open the Scenes panel.
Click Save Scene or similar.
Give it a descriptive name (e.g., “High-Value Customers in 2024”, “Fraud Ring Cluster”).

The scene stores:

Which nodes and relationships are visible.
Their layout and positions.
Applied styles, filter settings, and grouping (as far as supported).

You can later return to this scene to continue exploration or present it to others.

7.2 Share scenes and perspectives

Sharing methods depend on your environment and permissions, but typically you can:

Share Perspectives with other Bloom users connected to the same database so they see the same labels, styling, and search phrases.
Export scenes as images or screenshots for inclusion in reports, slide decks, or documentation.
Use public or shared Bloom workspaces (in Aeon or Aura-based environments, where available) to let others open the same graphical view.

This makes Bloom a powerful tool for cross-team collaboration, especially for:

Data analysts explaining findings to business stakeholders.
Architects showing graph models to developers.
Investigators or fraud teams outlining complex relationship patterns.

8. Use Bloom for specific visualization use cases

How you visualize data in Neo4j Bloom depends on your use case. Here are a few patterns to apply.

8.1 Network analysis and communities

For social networks, knowledge graphs, or communication networks:

Start from a seed node (e.g., a person or topic).
Expand relationships outward 2–3 hops.
Apply force-directed layout to see natural clusters.
Group or color nodes based on:
- Community or cluster label (if computed via algorithms).
- Role or category (e.g., customer vs. influencer).
Highlight central nodes (high degree, important hubs).

8.2 Fraud detection and investigations

For fraud, AML, or investigations:

Start from a suspicious account, transaction, or entity.
Expand targeted relationship types (e.g., TRANSFERRED_TO, OWNS, SHARES_ADDRESS_WITH).
Filter by time, amount, or risk score.
Use grouping (e.g., group by address or device) to see many-to-one relationships.
Save scenes showing:
- Rings of accounts around a central entity.
- Chains of transactions linking seemingly unrelated parties.

8.3 Customer 360 and product journeys

For customer analytics:

Search for a Customer by ID or name.
Expand relationships like PURCHASED, VIEWED, SUPPORT_TICKET, VISITED_PAGE.
Use time-based properties to understand sequences and journeys.
Color nodes by type (products, orders, tickets) to tell a clearer story.
Save scenes per customer segment or journey type.

9. Practical tips for effective Bloom visualizations

To make your Bloom graphs both beautiful and useful, keep these principles in mind:

Limit the initial result count
Always start small (e.g., a few dozen nodes), then expand selectively.
Keep Perspectives clean
Expose only the labels and relationships your audience needs; hide technical labels or internal relationships.
Use meaningful captions
Set captions to human-readable properties, like name, title, or sku, rather than internal IDs.
Color-code by concept
Use consistent colors across labels (e.g., blue for people, green for organizations, orange for events).
Use grouping for high-volume types
When there are hundreds or thousands of similar nodes (e.g., transactions), group them to avoid clutter.
Document scenes
Name scenes clearly and add notes (where possible) so others understand what they’re seeing.

10. Next steps for mastering visualization in Neo4j Bloom

To go further with visualizing data in Neo4j Bloom:

Experiment with different datasets:
- Create a temporary graph at https://sandbox.neo4j.com.
- Or spin up a free Aura database at https://console.neo4j.io.
Refine your Perspective:
- Add or hide labels.
- Tweak colors, icons, and captions until the graph “reads” naturally.
Combine Cypher and Bloom:
- Use Cypher to pre-select interesting subgraphs.
- Use Bloom’s visual tools to explore and present them.

By combining intuitive search, powerful styling, and interactive exploration, Neo4j Bloom becomes a central tool for answering the question “what’s really going on in this graph?”—letting you visualize complex relationships in a way that’s both understandable and actionable.