Not a chain of blocks
Hedera uses hashgraph consensus — a gossip protocol where nodes share what they know and, crucially, share metadata about who told them what. From that the network derives a consistent ordering of events without mining or leader election.
The structure is a directed acyclic graph rather than a linear chain. There is no block interval to wait for and no reorganisation risk in the conventional sense; consensus timestamps are assigned and finality arrives in seconds.
This changes what an explorer shows. There are no blocks to browse in the familiar sense — there are transactions with consensus timestamps. HashScan presents it that way rather than manufacturing a block abstraction, which is the honest choice and initially disorienting if you arrived from Bitcoin or Ethereum.
The trade-off is governance. Consensus nodes are operated by a council of named organisations rather than being permissionless. Anyone can use the network; not anyone can run a consensus node. That is a deliberate design decision with real implications, and whether it is acceptable depends entirely on what you are using it for.
Services rather than contracts
Hedera exposes native services rather than expecting everything to be built in smart contracts, which is a genuinely different architecture.
The Consensus Service provides ordered, timestamped message streams on topics. Applications use it as a tamper-evident audit log without deploying any contract. HashScan shows topics and their messages directly.
The Token Service creates and transfers tokens at protocol level. Like Cardano's native assets, these are not contracts — which means no contract to audit, no upgradeable transfer logic, and no approval exploit surface.
There is also an EVM-compatible smart contract service for Solidity, which sits somewhat awkwardly beside the native services and is the least distinctive part of the platform.
For an explorer, the interesting consequence is that most Hedera activity is legible without decoding. A token transfer is a token transfer at protocol level, not a contract call that must be interpreted.
Mirror nodes: the API question, solved differently
Hedera separates consensus nodes from mirror nodes. Consensus nodes run the network; mirror nodes maintain a queryable history and serve a public REST API.
This is an unusually sensible arrangement. The API is part of the platform rather than a product some company built on top, it is openly documented, and public mirror nodes serve it without a key. We confirmed this — no registration, no token.
It means HashScan is a front end over an API anyone can call, and anyone can run their own mirror node if they want independence. That combination — open API, open explorer, self-hostable infrastructure — is rarer than it should be, and it stands in contrast to the direction Etherscan and Blockscout have both moved in.
Our API comparison lists Hedera's mirror node among the genuinely keyless options, which is a short list.
Reading Hedera accounts
Hedera account IDs look like 0.0.123456 — a shard, a realm and an account number — rather than a hash. Readable, and initially odd.
Fees are denominated in US dollars and paid in HBAR at the current rate, which makes costs predictable in a way variable gas markets are not. HashScan shows both the HBAR amount and its dollar equivalent, which is the point of the design.
Token association is the mechanism that catches people: an account must associate with a token before it can receive it, which is the same structural idea as XRP Ledger trust lines and Stellar trustlines. The consequence is the same too — unsolicited token spam cannot happen, and a transfer to an unassociated account fails.
That pattern recurring across three unrelated chains is worth noting. Every network that requires explicit opt-in to hold an asset has eliminated token spam entirely, while every network that does not has a serious problem with it.
What our reference tests showed
We tested Hedera against items chosen to exercise its native services rather than its EVM compatibility layer, since the native services are the distinctive part.
An HBAR transfer rendered with both the HBAR amount and its dollar equivalent, which is the point of Hedera's dollar-denominated fee model and is presented well.
A Token Service transfer to an associated account worked and read cleanly. A transfer to an unassociated account failed, as the model requires, and the failure reason was legible — which is the same structural pattern as XRP Ledger trust lines and Stellar trustlines, and the same one we tested for on those chains.
The Consensus Service topic was the test with no equivalent elsewhere. We looked at an active topic and its message stream, each message carrying a consensus timestamp and sequence number. For an application using Hedera as an audit log rather than as a value ledger, that view is the entire product, and no general-purpose explorer would show it at all.
We confirmed the mirror node API works without a key from a clean request, which is the claim most worth verifying given how many "free" APIs now require registration. It did, and the response was well-formed and documented.
The EVM contract layer worked but felt like a guest. Contract calls rendered adequately and without the depth an EVM-native explorer provides. Our reading is that Hedera's own services are where this platform is coherent, and the Solidity compatibility is a bridge rather than a destination.