A validator in one sentence
A validator is a piece of software, running on a real computer somewhere, that participates in producing and verifying new blocks on a proof-of-stake blockchain. To get the job, the operator has to lock a deposit of the network's coin — a stake — that the protocol can take away if they misbehave. That economic skin in the game is how the network stays honest without mining.
If you are coming from Bitcoin or the early Ethereum era, the mental model to update is this: where proof of work spent electricity to make cheating expensive, proof of stake makes cheating expensive by putting capital at risk. The validator is the actor with that capital. For a complete picture of the consensus method itself, our what is proof of stake guide goes deeper.
What a validator actually does, hour by hour
On Ethereum, the canonical proof-of-stake chain, the work falls into a few quiet, repetitive tasks.
Attesting to blocks
Most of the time a validator is not proposing anything new — it is reviewing the blocks its peers just produced and signing off on them. This signature, called an attestation, says I have checked this block and I think it is valid. Attestations are how the network reaches agreement that yes, this is the canonical chain. They happen every twelve seconds and a validator is expected to participate in almost all of them.
Proposing a block
Every so often — roughly once every couple of months for a single 32 ETH validator on Ethereum — the protocol picks your validator to propose the next block. You assemble it from the mempool, broadcast it and earn the proposer reward plus tips from the included transactions. This is the more lucrative half of the job, even though it is the rarer half.
Staying online
A validator that goes offline does not get slashed (that is reserved for active cheating), but it earns nothing during the downtime and pays a small inactivity penalty if the network is struggling. The expected uptime is high. If you cannot keep a machine reliably connected, this is the practical reason most people delegate instead of running their own.
The stake — capital as security
The defining property of a validator is the deposit. On Ethereum the minimum is exactly 32 ETH. On Cosmos chains it varies. On Solana there is no fixed minimum — but in practice you need enough to be picked into the active validator set, which is competitive.
The stake serves two roles at once. First, it gates entry — you cannot become a validator without putting up real value, so attacking the network requires acquiring and locking a meaningful amount of the coin. Second, it is the collateral that can be confiscated if you cheat. Both effects work in the same direction: the network is secured by the validators' own money.
Stake also drives selection. The protocol picks block proposers and committees in proportion to how much each validator has staked. A validator with twice the stake is roughly twice as likely to be picked. This is unlike mining, where compute power decides — and unlike one person one vote, where the user count decides.
Slashing: the part you must understand
This is the rule that makes proof of stake actually work. If a validator does something the protocol can prove is malicious — most importantly, signing two different blocks at the same height (called equivocation) — the network slashes their stake. Part of the deposit is burned outright; the rest is gradually penalised over a few weeks; and the validator is ejected from the active set.
The penalties are intentionally severe so that the math never works out. To compromise consensus an attacker would need to control roughly a third of the active stake. Doing so would require buying enormous amounts of the coin (price would skyrocket as they buy), getting it past withdrawal queues, and then knowingly losing all of it to slashing. The defence is not technical; it is economic.
Less dramatic offences (going offline during a high-load period, missing too many attestations) result in small drip penalties. These are not slashing in the strict sense — they are inactivity leaks — but they hurt the validator's effective yield and exist to push operators toward good uptime.
Solo validating vs delegating
You have three realistic options if you want exposure to validator rewards.
Run your own validator
This is the maximalist path. You buy enough of the coin to meet the minimum, set up a node and an attached signing key, and operate it yourself. The yield goes entirely to you, you contribute to network decentralisation, and you are not trusting anyone. The trade-off is operational — you need stable internet, hardware that does not die, and the discipline to keep the software updated. A botched key setup or a cloud accident can cost real money.
Delegate to a professional operator
You hand your coins (or, more often, just the staking rights) to an operator who runs the validator on your behalf and shares the rewards minus a commission. The protocol still tracks the stake under your control; the operator can be slashed for misbehaviour but cannot run away with your principal. This is how most retail-sized stakers actually participate.
Hold a liquid staking token
Instead of locking your coins directly, you deposit them with a protocol that pools many users together, runs validators with that pool, and issues you a transferable token representing your share. The token earns staking yield while staying liquid — you can swap or use it in DeFi without unbonding. The risk is the protocol itself: a bug or governance failure in the liquid staking layer is a different threat than slashing on the base chain.
Validators vs miners — the cleanest comparison
It is worth being precise about what changed when Ethereum moved from proof of work to proof of stake in 2022.
A miner on a proof-of-work chain does the same job as a validator on a proof-of-stake chain — both produce blocks and earn rewards. The difference is the resource consumed. Miners burn electricity (and depreciate hardware) to win a hash race; validators lock capital and accept slashing risk. Both models work — they have just different cost structures and different attack surfaces.
Some practical consequences: proof-of-stake validators have no economies of scale at the hardware level (a 32 ETH validator runs on a tiny machine), but they do have one at the operational level (one team can run thousands of validators). Proof-of-work miners scale with capital expenditure into industrial farms. The decentralisation properties land in different places.
Bitcoin remains proof of work and is unlikely to change. Most newer chains are proof of stake. Both will probably persist for a long time.
What can go wrong for a validator
The honest list of risks for anyone running, delegating to, or using a validator.
Slashing. The largest tail risk. A misconfigured failover that signs two blocks at once will get you slashed even though you were not trying to attack anything. This is why solo validators avoid hot-spare setups.
Operator failure. Delegated stake is only as safe as the operator. A poorly run validator that gets slashed costs you too. Spread delegation across multiple operators if you can.
Smart-contract risk on liquid staking. If you hold a liquid staking token, a bug in that protocol can affect your principal even if the validators themselves are fine.
Withdrawal queues. Stake is not instantly liquid. Each chain has its own unbonding period, ranging from days to weeks. If you need cash quickly, this is a constraint.
Following the news that moves staking yields
Validator economics are not a constant. They move with issuance schedules, network upgrades, protocol-level fee burn changes, and shifts in how much of the supply is staked. A change in Ethereum's issuance curve or a Solana upgrade that compresses fees changes what every staker earns. Zippfeed surfaces those stories with a sentiment score (bullish, neutral or bearish) and an importance rating, so you can see which network-level news actually affects your validator's yield versus what is just noise. This is education, not financial advice — but the people earning the most consistent staking returns are the ones who notice these shifts early.
