Sending ADA, delegating stake and interacting with a Cardano application may look like three unrelated actions. Underneath, each depends on the same ledger rules: transactions consume existing outputs, stake pools order blocks through Ouroboros and nodes validate the result before it becomes part of the chain.
That combination is more useful for understanding Cardano than its old five-stage roadmap. The network has moved through several ledger eras since its 2017 launch, added smart contracts and entered community-led onchain governance. Cardano is still evolving, so this guide separates features running today from future work.
What Cardano is
Cardano is a public blockchain that uses proof of stake. ADA is its native asset: it pays transaction fees, represents stake in consensus and carries voting power in governance.
The project is associated with three founding organisations—Input Output, the Cardano Foundation and EMURGO—but that history should not be confused with the mechanism that now changes the protocol. The official eras overview records a sequence of ledger upgrades, including Alonzo for Plutus smart contracts, Babbage-era improvements and the Conway governance era. Chang introduced the initial governance framework; the Plomin hard fork in January 2025 enabled the full set of CIP-1694 governance actions.
A Cardano transaction, from output to output
Cardano uses an extended unspent transaction output model, normally shortened to eUTXO. A wallet does not simply subtract a number from one global account balance. It selects one or more unspent outputs as inputs, creates new outputs and returns any change to an address controlled by the sender.
Cardano extends Bitcoin’s UTXO design by allowing outputs to carry data and be locked by scripts. The eUTXO documentation explains an important consequence: validation depends on the transaction and the inputs it consumes. A wallet or application can evaluate a transaction before submission, although it can still fail if another transaction consumes an expected input first.
Fees are derived from protocol parameters, transaction size and, when scripts run, execution resources. They are not a fixed universal price. A transaction can transfer ADA, carry native assets, register a staking credential or interact with Plutus code. For a broader explanation of onchain programs and their failure modes, see our smart-contract guide.
How Ouroboros and staking fit together
Ouroboros divides time into epochs and slots. Stake pools can be selected to produce blocks, with selection influenced by the stake delegated to them. Other nodes verify the blocks and apply chain-selection rules. Cardano’s consensus documentation describes leader checks, block verification and chain selection as distinct responsibilities; “proof of stake” is therefore more than a rewards feature.
ADA holders may delegate voting power for block production without transferring ownership of their coins to a pool. Delegation also does not guarantee a fixed yield. Rewards vary with pool performance, saturation, fees, margin, pledge and network parameters. They come from transaction fees and controlled monetary expansion from the reserve. Any percentage shown by a wallet or calculator is an estimate, not a contractual return.
Plutus, native assets and application risk
Plutus is Cardano’s native smart-contract platform. Its onchain scripts decide whether a transaction may spend a particular output or perform another authorised action. Native assets can also exist directly in the ledger under a minting policy rather than requiring every token to reproduce a full fungible-token contract.
These design choices do not make an application safe by default. A Plutus script can contain flawed logic; an interface can send a user to the wrong transaction; governance or administrative keys may retain control; an oracle or offchain service can fail. Users need to inspect the application, token policy and transaction they are signing, not infer safety from the base network’s research process.
Governance after Chang and Plomin
Cardano’s governance is no longer a roadmap promise. Under CIP-1694, governance actions can involve delegated representatives (DReps), stake pool operators and a constitutional committee. Which bodies vote depends on the action. ADA holders can represent themselves or delegate voting power to a DRep, separately from delegating stake for block production.
The governance overview covers protocol changes, treasury withdrawals and constitutional checks. It also exposes a trade-off: stake-weighted voting gives ADA an operational role, but participation, delegation concentration and voter attention influence how representative the process is.
Scaling and remaining trade-offs
Cardano scales through improvements to the base protocol and through additional systems. Hydra, for example, is a family of layer-2 protocols whose first implementation uses multi-party state channels. A Hydra Head is useful for particular groups and transaction patterns; it is not evidence that every Cardano application automatically receives unlimited throughput.
The network’s main trade-offs are practical rather than ceremonial: eUTXO applications must manage input contention and transaction construction; developers work with a different model from account-based chains; stake and governance power can concentrate; bridges and applications add risks beyond the base ledger; and future roadmap items remain proposals until deployed.
How to evaluate a Cardano claim
Start by identifying the layer involved. Is the claim about Ouroboros consensus, the Cardano ledger, a Plutus application, a layer-2 system or an organisation building in the ecosystem? Then check whether the feature is active on mainnet, which authority can change it and what a failed transaction or compromised application would expose.
Cardano’s defining feature is not that peer review eliminates risk. It is the attempt to connect formal research, a UTXO-derived ledger, proof-of-stake block production and explicit governance. The strength of any specific use case still depends on the deployed code, operators, liquidity and choices made by its users.
Editorial note: this guide was fully reviewed and rewritten on September 3, 2026.
