The question that decides it is who you do not trust
Almost every blockchain brief we receive describes a database problem. The exception is narrow, it is worth naming precisely, and naming it is cheaper than building the wrong one.
A database, unless two or more parties who will not accept each other's records need to write to the same history. That is the entire test, and if one organisation controls the data, a ledger adds cost and removes nothing.
The test in one sentence
Ask who writes, and ask whether any writer would refuse to accept a record kept by another. If every writer reports to the same board, you have a database problem with a permissions question attached.
This is not a technical judgement. It is a question about the relationships between the organisations involved, and it is usually answerable in a meeting rather than a discovery phase.
What a ledger costs you
Latency measured in seconds rather than milliseconds. Immutability, which sounds like a feature until somebody enters a wrong figure. Key management, which is an operational burden you did not previously carry. And a much smaller pool of people who can maintain it.
Those are worth paying when the trust problem is real. They are pure loss when it is not.
- Writes measured in seconds
- Mistakes that cannot be edited, only appended to
- Keys somebody has to hold, rotate and recover
- A hiring pool an order of magnitude smaller
Where it does earn its place
Shared records between parties with no common owner: a consortium, a supply chain crossing companies, a settlement layer between institutions. In those the cost buys something a database cannot offer at any price, which is a history none of the participants can quietly revise.
What to take from this
- 01One owner means a database, every time
- 02The test is about trust between parties, not about technology
- 03Immutability is a liability until it is a requirement
- 04Answer this before scoping anything, not during
Ask the question directly
- What if we want it for the credibility?
- Then the requirement is marketing, and it should be costed as marketing. We will say so rather than build it, which is why some of these conversations end early.
- Is a private chain a middle ground?
- Rarely. A private chain with one administrator is a slow database with extra ceremony, and it usually satisfies nobody once the novelty passes.
Related answers
Blockchain development
Smart contracts, tokenised assets and shared records between parties who do not trust each other's database. We say first whether you need one, and usually you do not.
SEO or Google Ads first?
Bring the decision you are stuck on
A call, forty five minutes, no deck. We will tell you if we are the wrong firm.