We build blockchains, and the first thing we often tell clients is that they do not need one. A well-run database is faster, cheaper and easier to operate. A blockchain only earns its cost in one specific situation.
Answer these five questions honestly.
1. Do several organizations need to write to the same records?
If only your company writes the data, use a database. A blockchain is for records that several parties create together, such as shipments passing between suppliers, or claims shared between insurers.
2. Do they distrust each other, or a central operator?
If everyone is happy for one party (or a trusted vendor) to run the database, let them. A ledger helps when no single party should be able to change history on its own.
3. Does anyone need to prove later that a record was not changed?
Auditors, regulators and courts often ask "how do we know this record is the original?" A ledger answers with cryptography: every record is linked to the ones before it, so a change anywhere is visible to everyone.
4. Can you name who runs the nodes?
A private or consortium ledger needs each organization to run its own copy, called a node. If no partner is willing to run one, the system is a database with extra steps.
5. Is a cryptocurrency actually needed?
Almost never, for business use. A consortium ledger can use proof of authority: a fixed list of known organizations take turns adding blocks, and no tokens or mining are involved.
How to read your answers
- Yes to questions 1 to 4: a consortium ledger is worth a closer look.
- Yes only to question 3: you may just need tamper-evident logs, or anchoring record fingerprints to a ledger, which is much smaller.
- Mostly no: use a database with good access control and audit logs.
Axieo no longer takes on blockchain projects, but the habit behind this post (answer the hard questions first, in writing) is the same one behind every audit. See how we audit.