Is a full decentralised blockchain loyalty program possible?
Could it deliver additional benefits compared a centralised blockchain loyalty program with a decentralised cryptotoken?
Firstly, it’s important to clarify what is meant by a decentralised program. In a paper titled The General Theory of Decentralized Applications, dApps, by Johnston et al1, a decentralised application must meet four key criteria:
1. It must be completely open source, operate autonomously and with no entity controlling the token majority. Changes to the application must be adopted by consensus.
2. Data must be cryptographically stored in a distributed blockchain to avoid central failure points.
3. The application must use a cryptographic token for access to the application and as a reward to network supporters.
4. The application must generate tokens according to an algorithm that values contributions to the system.
For a loyalty program using a hybrid solution comprised of a non-blockchain loyalty platform connected to a blockchain platform, the application is clearly not decentralised. While a cryptographic token is used to reward supporters (for their engagement) and the application generates tokens that value contributions to the system (such as retail partner transactions), the loyalty program code isn’t open source, it doesn’t operate autonomously,
changes don’t require consensus and for the most part the data isn’t stored on a blockchain. There is also an entity controlling the token majority.
So, what might a fully decentralised blockchain loyalty program look like? This would involve development of the platform and running of the program to be jointly managed by a disparate group of individuals which evolve the program by consensus. Is this even feasible for a loyalty program, which requires extensive non-development support, such as retail partner negotiations and invoicing, marketing and member support to function at a basic level? And what additional benefits would it deliver?
According to Vitalik Buterin,2 developer of Ethereum, there are three arguments for why decentralisation is useful:
1. Fault tolerance: decentralised systems are less likely to fail accidentally because they rely on many separate components that are not likely (to fail simultaneously).
2. Attack resistance: decentralised systems are more expensive to attack and destroy or manipulate because they lack sensitive central points that can be attacked at much lower cost than the economic size of the surrounding system.
3. Collusion resistance: it is much harder for participants in decentralised systems to collude to act in ways that benefit them at the expense of other participants, whereas the leaderships of corporations and governments collude in ways that benefit themselves but harm less well coordinated citizens, customers, employees and the general public all the time.
The first two points make some sense from a system security point of view but there are other ways to protect core systems from attack which don’t require a decentralised approach.
The third point is much more relevant for a modern loyalty program. If we consider that major coalition loyalty programs have actively devalued their own currencies in order to boost their profits at the expense of their highly loyal members, we can see an argument for collusion resistance. Of course, this issue is already solved by replacing points and miles with a cryptotoken, as the value of the cryptotoken is decentralised. Another technique coalition loyalty programs have used to boost profits is adjusted their rules to increase the number of points which expire. For example, one major frequent flyer program’s terms and conditions stated members needed to earn at least one point or redeem at least one point every 24 months for all the points in the member’s account to remain active. With an increase in program engagement due to a rise in popularity for their program, the percentage of points expiring each year started to fall, so they reduced the number of months of inactivity from 24 to 18 months to boost their point expiry percentage to maintain their profitability. Under a decentralised loyalty program structure this type of rule change would require consensus. Once again this is already solved automatically by using a cryptotoken, as they never expire and therefore don’t require expiry rules. Thus, even the concept of collusion resistance doesn’t appear to be sufficient to argue for a fully decentralised loyalty program over a hybrid program.
One key argument put forward by industry commentators on the benefits decentralisation through blockchain can bring to the loyalty industry is the issue of fragmentation. The premise of the argument is that a single individual can belong to many loyalty programs,
each with their own currencies, rules and restrictions. This has led to a situation where tens of billions of dollars of points and miles are held in different accounts but at such a low value they cannot be used. By connecting all of these programs to a single blockchain program, members would be able to transfer them into a single token and bundle the value together, thereby unlocking the value and making loyalty great again. Businesses will be delighted by this advancement, the argument stresses, because it will remove a massive liability from their accounts.
Unfortunately, this utopian view of the future of blockchain loyalty is unlikely to ever eventuate, as the argument is built on misleading information. Firstly, most companies running their own loyalty program are doing so because they want to. If they wished to allow their members to pool their points or miles earned from multiple loyalty programs, they could have simply joined one of the many dominant coalition programs which have existed since the 1980s. They haven’t, either because they wish to keep the points and miles spend within their own ecosystem (thus ensuring the revenue doesn’t leave their walled garden) or because competitors have already joined the coalition program and locked them out with exclusivity provisions in the partner contract. In the case of the second point, they certainly wouldn’t want their members transferring their points or miles into tokens and pooling it with reward currency earned from their competitors, as there’s no competitive advantage for doing so.
With respect to balance sheet liability, it needs to be stated that this isn’t an issue for a well-structured loyalty program. An often cited, but in my opinion highly misleading, Harvard Business Review article by Kowalewski et al3 states ‘Blockchain could help relieve a large
balance-sheet liability that many in the industry are facing… Loyalty programs have long relied on cobranded cards and partnerships to sell points and generate incremental revenue. But the number of airline seats and hotel rooms available for redemption in recent years has been limited by near-record occupancy and load factors. The result has been a growing volume of unredeemed points, which new accounting standards have turned into a headache: Revenue attributable to the value of loyalty points must be deferred until the miles are redeemed.’
This argument is naive at best. The accounting standards governing points and miles based loyalty programs have been around for decades and they aren’t a headache and are relatively simple. When points or miles are awarded to a member, the loyalty program must defer enough revenue to a separate account to cover the cost of the future liability. As stated earlier, some programs, such as the major frequent flyer programs, are carrying liabilities of several billion dollars. Do they care? Not in the slightest. In fact, most of them have specific strategies to grow the liability over time. They wish for the number of points or miles earned each year to be slightly higher than the amount redeemed. By following their strict accounting processes, they have enough funds to cover the liability, therefore it isn’t an issue. In the meantime, they earn interest on the deferred revenue which adds to the program profitability.
In the event that a loyalty program allows for their currency to be redeemed on products outside their own ecosystem, they generally tend to structure their rewards range so the best value for members is to redeem on their own products (such as flights), with very poor values provided for third party products (such as gift cards and iPhones) in order to maintain most of the spend within their business. Some programs also allow their currency to be
transferred into the currency of another program. When this happens, the conversion rates are generally very poor to dissuade members from doing it. To argue that a majority of companies will support a unifying blockchain loyalty approach which allows members to transfer their points or miles out of their ecosystem and maintain the value is pure fantasy. It should also be pointed out that most companies promoting this idea as the central tenet of their business model are running hybrid platform solutions.
While I am not opposed to the idea of a fully decentralised blockchain loyalty program and continue to explore its potential, I’m yet to identify a meaningful benefit which cannot be delivered by a hybrid solution which utilises a standard loyalty design with a cryptocurrency instead of points and miles. I leave this discussion open for debate in the hope I may discover something which has thus far eluded me and will likely revisit the topic in future editions of Blockchain Loyalty and on www.blockchainloyalty.io
This is an excerpt from Blockchain Loyalty: Disrupting loyalty and reinventing marketing using cryptocurrencies (1st Edition) by Philip Shelper.
Blockchain Loyalty is available at all good book stores. Buy it now.
Philip Shelper is a specialist loyalty consultant based in Sydney, Australia who obsesses about everything to do with loyalty and rewards. His company Loyalty & Reward Co are a leading loyalty consulting firm.
Phil is the author of “Blockchain Loyalty: Disrupting loyalty and reinventing marketing using cryptocurrencies.”
www.blockchainloyalty.io is a global resource centre for everything blockchain loyalty.
Let’s connect! LinkedIn, Twitter