Blockchain: does your business really need one?


Tari Labs' Cayle Sharrock: You have to value what blockchain gives you to justify the cost of running one.

It’s 2020, and we’re still in hype overdrive about Blockchain. If conventional wisdom is to be believed, Blockchain is going to revolutionise and disrupt every industry known to humankind. It’s single-handedly going to redefine the traditional CFO role and revolutionise the finance function. And so on.

But here’s the thing. Blockchain probably isn’t the right solution for most of the problems it’s being touted to solve. Before CFOs start trying to find ways to use Blockchain to their advantage, they should start by asking if their industry actually needs it.

Fact is, blockchains are not terribly exciting on their own. At the end of the day, they are nothing more than databases. Really slow, inefficient and mind-bogglingly expensivedatabases, compared to the SQL clusters that sit in the IT departments of 90 of the world’s businesses.

For Blockchain technology to make sense to your business or industry, the benefits must outweigh these costs. You really have to value what a blockchain gives you to justify the immense cost of running one: tamper-evident logs (the block chain); cryptographic proof of ownership (digital signatures); public accountability (the distributed public ledger); and corruption resistance (proof of work).

So, using these four features as a checklist, let’s evaluate the banking and supply chain industries as possible use cases of blockchain technology, and decide whether the potential for CFOs is genuine, or whether it’s just buzzword bingo.

We’ve seen plenty of headlines over the past few years proclaiming how banks will use Blockchain to disrupt the industry. Usually, what they claim is that they can perform interbank settlements at a fraction of the cost of what the incumbent monopoly, SWIFT, provides.

So does Blockchain work for the banking sector? Clearly, tamper detection of the transaction history is a must-have here. What about digital signatures and proof of ownership? Without a doubt. Multiple signatures? The more the merrier.

Bitcoin was conceived as trustless money – and with banks, we have a fairly small community that is heavily regulated, and that do actually trust each other to some degree. Essentially, banks use governments’ big stick instead of proof-of-work to keep everyone honest. This works most of the time. Except when it doesn’t. The 2008 financial crisis and the 2012 Cypriot haircuts are just two examples.

How about public accountability from distributed public records? No, public accountability has never been the banking sector’s strong suit. That means a banks’ ideal “blockchain” is just tamper detection, plus digital signatures. This sounds like a bunch of databases that have tightly controlled access along with strong cryptographic signatures.

Banks gave this non-Blockchain blockchain a name: Distributed Ledger Technology. And it’s pretty much what SWIFT already does.

Verdict: Do banks need Blockchain? Nah. They want a cheaper alternative to SWIFT.

Supply-chain management
We’re often told how blockchain technology is going to revolutionise the supply-chain management (SCM) industry. BHP Billiton was one of the first large companies to announce in 2016 that they were implementing Blockchain for their core sample supply chain. We’ve heard similar stories about the diamond industry.

Should SCM CFOs be sitting up and taking notice? Whether you think a proof-of-work Blockchain makes sense for SCM is really secondary to the challenge of The Oracle problem: blockchains are brilliant at letting you know when data in the system has been compromised. But they have zero sense whether that data is true or not.

The Oracle problem arises whenever you need to bring the concept of truth, or providence from the real world into a trustless system like Blockchain. How does the core sample data get onto the blockchain ledger? Does someone type it in? Do they ever make mistakes? Can
they be bribed to type in something else? If it’s a totally automated system, can it fail? Be hacked?

Maybe we solve this by having two systems running and we compare the results. Or three. Or four. Now we have the problem of having to ship our samples to different labs around the world and be sure they weren’t tampered with in transit. If only we had a blockchain-based SCM system to secure our blockchain-based SCM system ...

Verdict: The Oracle problem is really hard, and torpedos a lot of tangible good-based blockchain proposals.

So, back to our original question: does your average CFO need a blockchain? Ultimately, the future of Blockchain applications (beyond money) lies in whether the benefits of having a decentralised, public record secured by proof-of-work outweighs its costs. There are plenty of really encouraging use cases emerging – think ticketing, for example, or trading in any digital assets. But for most industries, the jury’s still out.

Related articles