We need to talk about blockchain.

I think it's fair to say the blockchain hype has reached fever pitch. When companies can add the word ‘blockchain’ to their names and subsequently enjoy a 200 per cent uplift in their share value, one can probably infer that the market isn't operating with a great degree of rationality.

Much of this hype around blockchain has been driven by the recent wild and volatile fluctuations in the value of Bitcoin, which is underpinned by blockchain technology.

At the same time, an emerging market in enterprise blockchain has been making some noise. This market is immature and characterised by small-scale experimentation and pilots, with few significant projects operating at scale like IBM Global Financing.

Back in 2016, I wrote about the blockchain and its potential as a database for the networked society. My focus has always been on the transformative benefits new technology can deliver now and into the future.

But these days, I'm finding myself a little more sceptical of some of the claims being made for blockchain. That's not to say that blockchain isn’t an incredibly powerful and important tool – one that we're helping many of our own clients take advantage of. My point is that it’s not for everyone.

In gauging whether it’s right for your organisation, you need to start by scrutinising the problem you’re trying to solve. Some of these problems can be addressed with existing technologies or by tweaking current business models.

Take for example the concept of trust. In the case of public blockchains, such as the one that powers Bitcoin, trust is achieved through a combination of cryptography and a consensus mechanism.

Bitcoin ‘proof of work’ takes a game-based approach, where solving costly and time-consuming cryptographic problems produces data that’s used by people in the network to ensure agreement on the validity of transactions.

Other consensus mechanisms are available, but as we look to the permissioned blockchains many businesses are considering, the need to achieve trust through consensus in this way largely goes away.

As far as I can tell, enterprise blockchains revert to more traditional ways of establishing trust that are fundamentally based on reputation. In other words, you make mandatory decisions about whom to trust, and only grant permissions to read or write data on this basis.

If this is the case then the claims made for blockchain are largely equivalent to what’s happening with traditional database management software as it matures. We’re seeing greater cryptographic security, more and quicker synchronisation between trading partners, and more visible and complete data.

In this context, simply storing information on a permissioned blockchain doesn’t make it any more valid or true than data stored in a conventional database. In both cases, trust is a factor of reputation.

From this perspective, blockchain seems rather more like evolution than revolution. That pulls us back from blockchain’s decentralised model towards one where a centralised trusted authority remains important.

Of course, the ‘immutability’ that’s built into blockchain doesn’t come by default in more traditional approaches – an intermediary could modify data or change logs, for example. But if the data that’s stored on the blockchain isn’t valid and trusted from the outset, this becomes something of a moot point. Just because a piece of data is stored on a blockchain does not make it true – the old computing adage of ‘garbage in, garbage out’ still applies.

Having mentioned intermediaries, let’s consider another word that inevitably arises in any conversation involving blockchain – disintermediation.

Blockchain removes the intermediaries who sit at the centre of a business network or ecosystem and act as a single trusted authority for all the other participants (think banks, auditors, notaries, registries, etc.). Such trusted intermediaries are often criticised for causing friction and delay, and adding little value.

The point is, blockchain isn’t always the answer. We could, for example, achieve better outcomes using traditional models with networks organised in a slightly different way.

Central trusted authorities could maintain write privileges (the ability to enter or modify data), but members could communicate directly with one another, with the central authority having visibility of all communications between parties.

The database could effectively be completely open (equivalent to shared), but with read privileges (who can see what) determined by the needs of the participants. And multiple copies could still be made as necessary. Doing this should be technically feasible with existing database management technologies.

If we wish to implement business networks where trust is based on reputation, we may not necessarily need a blockchain if we think more creatively about how we can achieve the same thing with existing technologies.

Blockchain is an interesting and innovative technology. It's encouraging enterprises to think more deeply about how they manage trust. At the same time, organisations would be wise to be deliberate in how they source a solution that’s best for them.