Does your business need a blockchain? An honest test

Ask this: do several organizations, who don't fully trust each other, need one shared record that none of them controls alone? Think of a manufacturer, a shipping line, a customs broker and a bank, all updating the same shipment history, each worried the others might change it.

If the answer is no, you are done. A database with good access control and backups does the job, and your IT team already knows how to run one. A blockchain adds cost and complexity, so it has to buy you something a database cannot: a record that nobody can quietly edit, held jointly.

There are three realistic options, and for business use the middle one is usually the interesting one. A permissioned ledger is shared by named partners, each running a node, with rules about who can read and write. A public chain is open to anyone, which suits few business processes and raises more privacy and regulatory questions.

For almost every company we talk to, the honest answer is the first column. When the answer is the second, the group of partners and their agreement on rules matter more than the technology.

A shared ledger costs more than a database to build and to run, and it needs several companies to agree on rules, onboarding and who pays. That coordination is often harder than the code. If the benefit is small, the project stalls at the first meeting.

Privacy is the other trap. Data written to a ledger is hard to remove, which clashes with privacy laws that give people a right to deletion. The usual pattern is to keep the data in each party's own systems and record only a fingerprint, so anyone can prove a record hasn't changed without seeing it. Rules differ by country, so check them locally.

If your design includes smart contracts, programs that run on the ledger and enforce agreed rules, treat them like a bank's payment code. Once deployed they are difficult to change, and a bug can lock up or misroute real business transactions.

A production smart contract should be reviewed by an independent auditor who did not write it, tested thoroughly, and rolled out in stages. We build it that way and we would ask any vendor, including us, to show you the audit report before go-live.

Start small. Pick one process, two or three partners who already want it, and a clear measure of success such as fewer disputes or faster reconciliation. Keep it business-to-business, with no tokens for sale and no custody of anyone's funds.

Agree on the rules in writing first, build the smallest version that works, and only then think about expanding. If partners lose interest before launch, you have learned something cheaply.