A private blockchain is a ledger where only approved organizations run nodes and only approved identities can transact. It makes sense when several parties who do not fully trust each other need one shared, tamper-evident record, and none of them should own the database. If a single company controls all the data, a private blockchain is almost always the wrong tool.
Private, permissioned, consortium: the terms
People use these words loosely. A private network is operated by one organization, which mostly gives you an append-only database with extra steps. A consortium or permissioned network is run by several organizations, each operating nodes and sharing governance. Nearly every enterprise project that delivers real value is the second kind, because the benefit comes from removing the need to trust one operator.
The main platforms
| Platform | Model | Smart contracts | Good fit |
|---|---|---|---|
| Hyperledger Fabric | Channels, endorsement policies, ordering service (Raft) | Chaincode in Go, Java or JavaScript/TypeScript | Multi-party workflows needing fine-grained data partitioning |
| Hyperledger Besu | Ethereum client with permissioning and QBFT or IBFT 2.0 consensus | Solidity, full EVM | Teams wanting Ethereum tooling and a path to public networks |
| Corda (R3) | Point-to-point state sharing, notaries, no global broadcast | Kotlin or Java contracts and flows | Regulated finance where data must be shared only with deal participants |
| GoQuorum | Ethereum fork with private transactions via Tessera | Solidity | Existing deployments; ConsenSys has steered new users toward Besu |
Fabric and Besu are now projects of LF Decentralized Trust, the Linux Foundation umbrella that succeeded the Hyperledger Foundation in 2024. More detail on each lives in the guides to Hyperledger development, Corda and Quorum.
Designing the network
Governance before code
The questions that decide success are not technical: who may join, who approves new members, who pays for nodes, how contract upgrades get approved, what happens when a member leaves, and which jurisdiction's law governs disputes. Write these down in a consortium agreement before you choose a platform.
Consensus
Permissioned networks use voting-based Byzantine fault tolerant protocols, not mining. QBFT in Besu tolerates fewer than one third of validators being faulty or malicious, so four validators survive one failure. Fabric's Raft ordering is crash fault tolerant only, which is acceptable when the orderers are run by trusted parties but worth stating plainly to members.
Privacy
Every node seeing every transaction is a deal-breaker in most business settings. Options include Fabric channels and private data collections, Corda's need-to-know model, Besu privacy groups, or storing only hashes on-chain and keeping documents in each party's own systems. The hash-anchoring pattern is simple and underrated.
Identity and keys
Permissioned chains rely on a PKI. Plan certificate authorities per organization, hardware security modules for signing keys, rotation, and revocation when an employee leaves.
Integration is most of the work
The ledger is rarely more than a quarter of the effort. The rest is connecting ERP, document systems and identity providers, writing event listeners that push ledger updates into existing workflows, building dashboards, and operating the nodes with monitoring, backups and upgrade procedures across organizations that run different IT stacks.
When a database is better
- One organization controls the data and everyone already trusts it.
- You need to delete or correct records freely; GDPR erasure requests are awkward on immutable ledgers, which is another reason to keep personal data off-chain.
- Throughput and latency requirements are tight and there is no multi-party trust problem.
- Participants will not commit to running nodes. A "consortium" where one vendor hosts every node is a database with a blockchain label.
A short, scoped proof of concept is the cheapest way to test whether the multi-party benefit is real before committing a consortium to production.
Realistic timelines
A focused pilot with two or three members, one workflow and a single platform typically takes three to six months, much of it spent on governance and integration. Production rollouts across many organizations commonly take a year or more. If someone promises a production consortium in weeks, ask who has signed the governance agreement.
For a wider view of enterprise options, see enterprise blockchain solutions.
Frequently asked questions
Is a private blockchain more secure than a public one?
Not automatically. It is more controlled: fewer, known validators and restricted access. Security depends on how keys are managed and how many independent parties run validators. A network where one company controls a majority of validators offers little tamper resistance against that company.
Do private blockchains have tokens or gas fees?
Usually not in any economic sense. Besu can run with zero gas price, and Fabric has no native token. Tokens appear only when the use case itself involves a digital asset, such as tokenized deposits.
Can a private network connect to public blockchains later?
EVM-based networks like Besu make it easier because contracts and tools are compatible. Teams often anchor periodic state hashes to a public chain for extra assurance, or bridge specific assets when regulation allows.
Which platform should we choose?
Pick Besu if your team knows Ethereum or you may go public later, Fabric for complex multi-party data partitioning, and Corda for regulated financial agreements shared only between counterparties. Team skills and member preferences often decide it.