When a large transfer, a hack, or a suspicious withdrawal happens on a public chain, the natural next question is “where did the money go?” Because the ledger is public, you can often follow it yourself, hop by hop, using nothing more than a block explorer and patience. This guide covers the actual technique and, just as importantly, where it stops working.
Starting from a known transaction or address
Tracing starts from something concrete: a transaction hash tied to the event you’re investigating, or an address already identified as belonging to a specific party (an exchange hot wallet, a project treasury, an attacker’s first-receiving address). From there, the method is mechanical: open that address’s page, look at its outgoing transactions after the event in question, and follow the funds to the next address, repeating the process.
Step by step: following a trail
- Open the starting address or transaction and note the exact amount and asset moved, and the timestamp.
- Look at the receiving address’s own transaction history for what it did with the funds shortly after receiving them – did it hold them, split them across multiple outgoing transactions, or forward them in one hop to a single new address?
- Repeat the process at each new address, tracking amounts carefully since funds are frequently split, combined with other funds, or partially moved while a remainder stays put.
- Watch for addresses that an explorer or a community-maintained label set already identifies as something meaningful – a known exchange deposit address, a known mixer or bridge contract – since reaching one of these is often where a simple hop-by-hop trace either ends cleanly (an exchange deposit, which law enforcement or the exchange itself can potentially act on) or effectively goes dark (a privacy-preserving mixing service, after which the original coins can no longer be distinguished from others pooled alongside them).
- Note every hop’s transaction hash as you go, since a trace is only as useful as a reader’s ability to independently re-walk the same path and confirm it.
Heuristics used on UTXO chains, and why they differ by chain design
The specific heuristics available depend on how a chain structures transactions. Bitcoin and other UTXO-based chains allow a single transaction to spend several separate inputs at once, which gives rise to a well-known heuristic: addresses used together as inputs to the same transaction are very likely controlled by the same entity, since spending them together generally requires holding the keys to all of them. Chain analysts also look for “peeling chains” on these chains, where a wallet repeatedly sends most of a balance onward through a sequence of transactions while keeping a small remainder behind at each step. Account-based chains like Ethereum work differently – a transaction has exactly one sender – so the common-input heuristic does not apply there in the same form; tracing on account-based chains instead leans more heavily on simple hop-following plus cross-referencing addresses against lists already identified through other investigations or voluntary disclosures. Either way, these are probabilistic groupings, not direct proof, and should be described as such rather than stated as certainties.
Where tracing stops proving anything
A hop-by-hop trace proves that specific amounts moved between specific addresses at specific times – that much is as solid as any other on-chain fact. It does not, on its own, prove who controls any of those addresses in the real world. It also reliably breaks down at certain points: cross-chain bridges move value to a different chain, sometimes through a design that doesn’t preserve a one-to-one on-chain link between the deposit and the corresponding withdrawal on the other side; privacy-focused mixing services are specifically built so that funds entering cannot be cryptographically linked to specific funds leaving; and an address simply going quiet for a long period doesn’t tell you whether its controller has stopped, is waiting, or has moved to tooling that doesn’t leave the same visible trail. Any write-up that claims to have “traced the funds to [a named person]” without a separate piece of real-world identification evidence – a KYC record disclosed through legal process, a voluntary admission, an exchange’s own confirmation – has gone further than the on-chain data alone supports.
Documenting a trace so someone else can check your work
A trace is only as credible as its reproducibility. The strongest write-ups list every hop’s transaction hash, in order, alongside the amount and timestamp at each step, so that anyone with access to the same public block explorer can independently re-walk the exact path and arrive at the same conclusion. A trace summarized only in prose, without the specific hashes that let someone else check it, is asking a reader to simply trust the summary, which defeats the entire point of working with public, independently checkable data in the first place.
It’s also worth explicitly noting where a trace involved judgment calls rather than purely mechanical hop-following – for example, choosing one output among several as the “main” continuation of a trail when a transaction split funds across multiple addresses. Being transparent about where a heuristic or an assumption was applied, rather than presenting every step as equally certain, is what separates a careful trace from one that overstates its own certainty.
Not financial advice
This guide explains a research technique for public blockchain data. It is general education, not financial, legal, or investment advice, and is not a substitute for professional chain-analysis tooling or law enforcement process where funds tracing has legal consequences.
When manual tracing becomes impractical
Hop-by-hop manual tracing through a block explorer works well for a small number of hops but becomes slow and error-prone once a trail splits repeatedly across many addresses. Dedicated chain-analysis tools exist specifically to visualize and query this kind of branching flow at scale, and professional investigators generally use them rather than manually clicking through dozens of addresses. Knowing the manual technique still matters, both for spot-checking a specific claim and for understanding what any tool’s visualization is actually built from underneath.
Why timestamps deserve as much attention as addresses
A trace that correctly follows the right addresses but glosses over timing can still mislead – funds moving through an address months apart tell a very different story than funds moving through in rapid succession, and a careless write-up that omits timestamps invites a reader to assume a tighter, more suspicious sequence than what the record actually shows. Recording the exact timestamp at every hop, not just the amount and address, keeps a trace honest about pacing as well as path.
FAQ
Can I trace funds across different blockchains?
Only indirectly. Moving value between chains usually goes through a bridge, and depending on the bridge’s design, there may or may not be a clean, directly linkable record connecting the deposit on one chain to the corresponding release on another.
Does reaching a mixing service mean the trail is permanently dead?
For a simple hop-by-hop trace, effectively yes – that is the specific purpose these services are designed to serve. Professional chain-analysis firms sometimes have additional, non-public techniques for certain mixing designs, but that’s a different and more specialized capability than what this guide covers.
If funds end up at a known exchange address, does that identify the person who moved them?
Not by itself, to an outside observer. It does mean the exchange, as a regulated custodian in many jurisdictions, likely holds know-your-customer information tied to the account that received the deposit, which is why reaching an exchange deposit address is often a meaningful endpoint for investigators with legal authority to request that information.
Is it safe to assume addresses used together in one transaction belong to the same person?
On UTXO chains like Bitcoin, where a transaction can spend several inputs at once, it is a reasonable, widely used heuristic, but it is still probabilistic rather than certain. It does not apply the same way on account-based chains like Ethereum, where a transaction has only one sender.
Not financial advice. Human-edited; AI-assisted drafting is disclosed in our editorial guidelines. Spotted an error? See how corrections work.
