Few technologies have been oversold as much as blockchain, and few are as useful when applied to the right problem. The question to ask is never "should we use blockchain?" It is "do we have the problem blockchain was built to solve?" This guide walks through how we answer that with clients before a line of code is written.
The problem blockchain actually solves
At its core, a blockchain is a shared record that several organisations can write to, where no single one of them controls the history. Every participant can check that what is recorded has not been changed after the fact.
That is valuable when several parties need to agree on the same facts (who owns what, what was delivered, what was paid) and none of them is willing, or allowed, to let one party keep the master copy.
Signs that blockchain is a good fit
- Several independent organisations need to read and write the same records.
- There is no trusted central party, or relying on one is expensive or slow.
- Participants need to prove, later, exactly what happened and when.
- Assets or rights need to move between parties with rules enforced automatically.
- Reconciliation between separate systems is a recurring cost today.
Signs that a normal database is the better choice
If one organisation owns the data and everyone already trusts it to run the system, a well-designed database will be faster, cheaper and simpler to change. Adding a blockchain there adds cost without removing a problem.
- A single company controls the data and the process.
- Records need to be edited or deleted freely, for example for privacy requests.
- Very high transaction volumes with millisecond latency are the priority.
- The only goal is to appear innovative.
Public or permissioned?
Public networks such as Ethereum are open to anyone and suit tokens, payments and applications that need to be verifiable by the public. Permissioned networks limit who can take part, which suits consortia of businesses that need privacy, known participants and governance agreed in advance.
The choice follows from who needs to verify the records and what may be seen by whom. We settle it early, because it shapes the architecture, the costs and the legal questions that follow.
What a sensible first project looks like
Start narrow: one process, a small number of participants and a clear measure of success, such as fewer disputes or less time spent reconciling. Integrate it with the systems people already use, so the ledger works in the background rather than asking everyone to change how they work.
The best blockchain projects are the ones where users never need to know a blockchain is involved.
Security is part of the design, not a final step
Smart contracts are hard to change once deployed, and they often hold value. Specification, testing, independent review and a plan for upgrades and key management belong in the project from the first week, not at the end.
Key takeaway
Blockchain is the right tool when several parties need one trusted record that none of them controls. If that is not your situation, you will get more from a well-built conventional system, and a good partner will tell you so.
Have a project like this in mind?
Talk to an engineer about it. Free consultation, with no obligation.
Get free consultation
