Does "1 million active addresses" mean 1 million users?
Because every transaction on a public blockchain is visible, anyone can count how many addresses transacted in a day and how much value moved. Figures like "one million active addresses" or "record daily volume" are often cited as evidence that a project is growing.
On-chain metrics need to be read with a clear idea of what they count. Addresses are not people, and transfer volume is not the same as economic activity. This guide explains what the main metrics measure, what inflates them, and how to cross-check them, using hypothetical data.
What on-chain metrics cover
On-chain metrics are calculated from transactions and state recorded on the blockchain: balances by address, transaction counts, tokens transferred, smart-contract calls, and total fees. Ethereum.org's page on data and analytics explains that block explorers, indexing services, and analytics tools can be used to query and process this data.
Activity that happens off-chain does not show up. Trades inside a centralized exchange are recorded only in the exchange's own ledger, so coins traded mostly on exchanges can look less active on-chain than they really are.
Addresses are not users
As Ethereum.org's page on accounts explains, an address is an identifier derived from a key pair, and anyone can create as many as they like. One person may use many addresses, while a single exchange or service may hold many customers' assets in a few. Address counts are only a rough proxy for users.
Creating thousands of addresses to farm airdrops, or bots repeating the same transaction pattern, can inflate active addresses dramatically. A sudden jump right after a rewards program launches is a reason to look for exactly these causes.
Hypothetical data compared
The table lays out one day of hypothetical network data by metric, with what can inflate each one and how to cross-check it.
| Metric | Hypothetical value | What it counts | What inflates it | Cross-check |
|---|---|---|---|---|
| Active addresses | 1 million | Addresses that took part in a transaction | Bots, multiple addresses farming airdrops | Per-address patterns, reward schedules |
| Transactions | 5 million | Transactions recorded in blocks | Automated repeats, small spam | Average fee, distribution of transaction sizes |
| Transfer volume | $2 billion | Value moved between addresses | Same funds moving repeatedly, internal transfers | Recalculate excluding known exchange and service addresses |
| Fee revenue | $500,000 | Actual cost paid by users | Relatively little | Whether activity carries real cost |
Fee revenue reflects money users actually spent, so it is harder to inflate. If active addresses and transactions jump while fees do not, the quality of that growth deserves questioning.
Reading activity and volume
Transfer volume adds up the value of tokens moving between addresses, so the same money counted several times inflates it. Exchanges moving funds between their own wallets, or DeFi contracts passing tokens internally, can also be included.
As a hypothetical example, a single $10 million swap that routes through five contracts can be recorded as $50 million of transfer volume. When reading volume, check whether internal movements between known exchange or bridge addresses are excluded and how the data provider defines the metric.
Checking collection methods
Providers define the same metric differently. Whether "active addresses" counts senders only or receivers too, and whether contract addresses are included, can change the number a great deal. The daily cut-off time and whether layer 2 activity is included also vary.
When comparing metrics from different sources, read each provider's definitions first, and where possible track changes within a single source's time series. For metrics published by a project itself, check whether it discloses the tools and definitions it used.
A useful habit is to keep a short note next to any metric you rely on: the provider, the definition, and the date you checked it. Definitions do change, and a series that suddenly jumps may simply reflect a methodology update rather than new activity.
An on-chain checklist
When reviewing a project's on-chain metrics, work through these steps.
- Read the provider's definition of each metric: what is counted and over what daily window.
- Look at changes in active addresses and transactions alongside fee revenue or other costly activity.
- Check whether spikes coincide with airdrops, reward programs, or bot activity.
- Confirm whether transfer volume excludes internal movements between known exchange and bridge addresses.
- Compare figures with another provider or with raw data from a block explorer instead of relying on one source.
Limits of on-chain analysis
On-chain data is transparent, but it does not reveal who owns an address or why a transaction was made. The same figures can support different interpretations, and a wrong reading can lead to a bad decision. The CFTC warns of manipulation and fraud risks in virtual currency markets.
The data here is hypothetical. When using real metrics, check the provider's current definitions and the underlying data with their dates, and look for several metrics pointing the same way rather than relying on one.