Nodes in the Transaction Graph
Each node is a transaction.
Some papers and tools also model each UTXO as a node instead (this is called the UTXO graph).
Edges in the graph
An edge represents the flow of value:
If a transaction A creates an output (UTXO), and a later transaction B uses that output as an input, you draw a directed edge from A → B.
This captures the fact that B depends on A.
Graph structure
It’s a directed acyclic graph (DAG):
Transactions only point forward in time (can’t spend future coins).
No cycles, because once an output is spent, it cannot be recreated.
All transaction paths ultimately lead back to:
Coinbase transactions (block rewards), which are the origin of all bitcoins.
What the graph shows
Flow of coins: You can trace any satoshi (or UTXO) back to its origin.
Clustering analysis: Because outputs can be linked when spent together, analysts can try to infer common ownership.
Network health: Graph analysis is used in forensics, compliance, and research.
- Example
Imagine:
- Tx1: Coinbase creates 50 BTC → Output A.
- Tx2: Uses A (50 BTC) as input → sends 30 BTC to Bob, 20 BTC back to Alice.
- Tx3: Uses Bob’s 30 BTC to pay Carol 10 BTC and himself 20 BTC.
The graph would look like:
In summary:
The transaction graph is a DAG connecting all Bitcoin transactions by their input/output relationships. It’s the backbone for tracing funds, analyzing usage patterns, and enforcing Bitcoin’s rule that each UTXO can be spent only once.