What Are Daily Active Users (DAU)? A Practical Guide

Daily active users (DAU) is the number of distinct users who meet a product’s definition of “active” during a specified day. A person who performs the qualifying action several times that day is generally counted once, not once per click or transaction.

The metric has two parts you need to check:

• Who counts as a user? It may be an account, device, browser, or—common in crypto—a wallet address.
• What counts as active? It could be opening an app, connecting a wallet, making a swap, supplying funds, or completing another chosen action.

The measurement window matters too. “A day” might mean a calendar day in a stated time zone or a rolling 24-hour period. Those approaches can produce different totals, especially around midnight.

DAU is a measure of activity, not automatically a measure of product quality, revenue, user satisfaction or unique human beings. A high number can indicate frequent use, but only when you know what the service counted and how it handled repeat activity, bots and identity.

In a conventional app, a product team may identify users through account logins, device identifiers or analytics tools. A crypto dapp often has no account registration, so its team may use wallet addresses or on-chain activity as a proxy for people. That proxy has limits: one person can use multiple wallets, and more than one person can share a wallet.

Also distinguish activity on the dapp’s website from activity on its smart contracts. Someone might visit a page without connecting a wallet; a transaction may be sent through an aggregator or another interface without visiting the dapp’s own site. On-chain records can show addresses and transactions, but they do not by themselves establish how many human users acted.

For example, a DAU figure for Uniswap could be based on distinct wallets swapping through specified contracts, while an app-analytics figure could count visitors to its interface. Those are different measurements. The same issue applies to Aave: a person who checks a position but does not transact may count in one measurement and not another.

Treat a crypto DAU number as a defined activity count, not a verified count of people, unless the project explains a stronger identity method.

You can understand a DAU calculation by writing down the definition before looking at the total. A basic formula is:

DAU = distinct qualifying users during the stated measurement period

To make that usable, specify each part:

1. Choose the user identifier. Examples include account ID or wallet address. Say whether addresses are deduplicated across networks.
2. Choose a qualifying event. Decide whether a visit, wallet connection, successful transaction or other action is enough.
3. Set the time window. State the date, time zone and whether the period is a calendar day or rolling 24 hours.
4. Count each identifier once. Multiple qualifying actions by the same identified user in that window should not inflate a unique-user count.
5. Document exclusions. Explain any filtering for internal testing, known bots, failed transactions or spam, and how that filtering works.

If a dashboard reports 8,000 active wallets, for instance, that describes distinct wallets under its rules—not necessarily 8,000 people. Without the event, identifier and time-window definitions, a DAU total is hard to reproduce and even harder to compare with another service.

When you see a DAU figure for a crypto project, look for its measurement rules before treating it as a benchmark. Ask whether the count represents unique wallets, signed-in accounts, site visitors or some combination. Then check whether it includes every network, only certain contracts, and activity from third-party interfaces.

A useful reading checklist is:

• Event: Does “active” mean a visit, connection, transaction, or a specific product action?
• Identity: Is the unit a wallet, account, device, or estimated person?
• Scope: Which app versions, chains and contracts are included?
• Period: What date and time zone does the number cover?
• Filtering: Are bots, failed calls, test activity or repeat wallets handled consistently?

If the project does not publish these details, do not assume them. Say that the DAU definition is unclear and avoid comparing the total directly with a service that provides a full methodology.

Also distinguish a single day’s count from an average over a period. “DAU” may refer to one daily value, while a chart may show daily values across a month. A monthly average can smooth peaks and dips, but it is not the same as the number of people active every day.

DAU is most useful as a trend measured consistently over time. If the same product counts the same kind of activity with the same rules each day, a rise or fall can signal a change worth investigating. It does not explain why that change happened. A promotion, a market event, a new feature, a bot campaign or a change in tracking could all affect the number.

Compare DAU with other measures that answer different questions:

Metric	What it helps describe
MAU	Distinct users active during a month
Transactions	How many actions were recorded, not how many users acted
Volume	The value or quantity moved, depending on the project’s definition
Retention	Whether users return after an initial visit or action

DAU divided by MAU is sometimes called a stickiness ratio. It gives a rough indication of how much monthly activity occurs on a typical day, but it is not a cohort-retention measure and can hide users who appear only once.

For a lending app such as Aave, transaction counts and supplied or borrowed balances tell you different things from unique active wallets. None of these metrics alone shows whether activity is useful, sustainable or safe for users.

A single DAU number can mislead when the counting rules or activity mix change. A burst of token incentives may attract short-term transactions without creating repeat use. Automated bots can produce many actions, while one person operating several wallets may appear as several users. Conversely, a shared wallet can make multiple people look like one.

Common mistakes to avoid:

• Treating addresses as people. Call them active wallets unless the methodology supports a person-level count.
• Comparing unlike totals. A wallet-transaction count is not directly comparable with logged-in app accounts.
• Reading transactions as users. One busy address can generate many transactions, so action volume can rise while DAU stays flat.
• Ignoring measurement changes. New chain coverage, contract updates or analytics changes can create a jump that is not organic growth.
• Assuming high DAU means a good product. Activity does not establish safety, user satisfaction, profitability or fair distribution.

Before drawing a conclusion from a sharp change, check whether the project changed its interface, incentives, supported networks or counting method around the same time. If the cause is not clear, describe the movement as a change in reported activity rather than proof of adoption or decline.

Use DAU as one input in a decision, not as a shortcut for judging a dapp. First establish what the metric counts. Then look for consistency across time and compare it with relevant measures such as retention, transaction activity, liquidity or service reliability. Which companion measures matter depends on what the product does.

For a wallet-connected exchange, check whether activity means a swap, a wallet connection or a site visit, and whether multiple networks are combined. For a lending protocol, look separately at users who supply, borrow or repay; a combined count can conceal very different behavior. Do not assume that a larger DAU automatically means deeper liquidity or lower risk.

A practical decision sequence:

1. Find the DAU definition, date and scope.
2. Identify whether the unit is people, accounts or wallet addresses.
3. Check for a consistent time series and note any methodology changes.
4. Compare with metrics that fit the product’s actual purpose.
5. Treat unexplained figures as uncertain, and make security and financial decisions using more than user counts.

If a project publishes only a headline number, you can still note it as a reported activity measure—but you cannot confidently infer how many distinct people use the product or why they use it.

A reliable DAU report should let you tell what was counted, when it was counted and how repeat activity was handled. Look for a clear definition of the user identifier, qualifying event, networks or product areas included, time zone, and any filtering process. A chart is more useful when the methodology stays stable and changes are disclosed.

Check that the reported total matches the claim being made. “Active wallets” is a reasonable label for a distinct-address count; “users” can imply people and deserves more explanation. If the project has no public methodology, avoid filling the gaps with assumptions. You can describe the number as reported DAU and state that its counting rules are not published.

For your own tracking, keep a short record alongside each figure:

• date and period covered;
• source of the count within the product’s own reporting;
• user identifier and active event;
• included networks or app surfaces;
• known filtering or methodology changes.

That small amount of context makes trends easier to interpret and prevents an apparent growth spike from being mistaken for a real increase in human users. It also makes comparisons fairer when you revisit the metric later.