Time to bid farewell to transaction processing

In years to come, we will scratch our heads and ask why anybody would have bothered with such a restrictive construct.

For more than half a century, IT business systems have been constructed around the notion of a discrete set of transactions that trigger some sort of activity. There are now clear indications that the day of the transaction is numbered, at least as we have come to know it.

Prior to business computing, people would need to line up at a bank, a shop or a government agency and wait to be served. The processing of each request was manual, so procedures were designed to minimise the number of interactions necessary to satisfy each customer request. Forms needed to be filled in and details had to be verified.

As computers began to take over manual processing, the notion of a transaction was a perfect fit for automation. Paper forms and manual processing could easily be translated into a discrete set of batch transactions. Later, as online systems replaced these clunky old systems, transactions became smaller and were processed more quickly, but the transaction still remained the cornerstone of IT service delivery. Service-level agreements were based on transaction response time and transaction throughput.

This is all very logical when viewed from an internal efficiency perspective, but is mind-numbingly bureaucratic when viewed from a customer perspective. There are simple alternatives.

For example, a growing number of tax agencies around the world are finding that a better alternative is to embed government tax processing into commonly used accounting software, rather than interacting directly with the client.

As another example, Ovum recently received an international enquiry from a government agency involved in public transport. It asked about the best way to create a transport timetable app. Our advice was that there is already a thriving ecosystem of transport apps that are well integrated into tourism, mapping and other outcome-specific solutions. Ovum advised it may be better to first look at leveraging that ecosystem rather than writing yet another single-purpose app.

Today, the simplicity and immediacy of mobile apps and social networks can provide opportunities to reassess the traditional transaction processing paradigm. Consider the last time you made an online purchase. The process has become much more sophisticated than just an order transaction. Typically, the vendor engages the client in a long-term conversation so that, over time, the vendor knows more and more about the client’s needs and preferences and can tailor services to meet them. Analytics is used to anticipate requirements and offer options and opportunities before the need arises. The conversation is used as an opportunity rather than an overhead.

Under this sort of model, government service delivery would be about anticipating needs and shaping community behaviour rather than enforcing legislation. It is all about driving better long-term outcomes, rather than just processing a transaction.

Image courtesy Quinn Dombrowski under CC