Recently, while reading through a project's retrospective notes, I was reminded of my very first assignment in my Computer Science course at the University of Manchester. It was an 800-word essay on why systems fail, focussing on the failures of the London Ambulance Service’s computer-aided dispatch system project in the late 90s.
Beynon-Davies (1999), who wrote a detailed paper on the case, concludes that software failure is multifaceted and cannot be attributed solely to the software development phase, which, even today, is often the first to be blamed. In fact, many factors, both before and after implementation, contribute equally to success or failure.
So, if systems fail that’s because projects fail.
Digital Projects Today
Digital projects have evolved significantly since the late 90s. Today, a digital system isn’t just about what it allows users to do but also how effortless and enjoyable it is to use. It’s about both functionality and user experience.
The infrastructure required to host digital systems has also transformed. In technical terms, where a few servers with manual deployments once sufficed, we now rely on containerized public clouds, distributed file systems, high-availability databases, multivariate caching, and autoscaling environments for development, integration, and deployment pipelines.
The scope of projects has expanded dramatically as well, now encompassing composable digital experience platforms with decoupled front-ends, complex system architectures, and numerous third-party integrations.
But this brings complications. It’s rare for a client to know everything a system should do upfront.
Especially when systems are meant to influence customer experience and expectations, they must be aligned with company goals, roadmaps, and budgets, all of which are dynamic by nature. These systems also need to be integrated with customer support systems, marketing activities, brand guidelines, tone of voice, and more.
It’s a People’s Game
Consider all the people who need to be involved before, during, and after a digital project. Who these individuals are influences the success or failure of the project. Factors include how skilled they are, how well they collaborate, how clearly they understand what needs to be delivered, and whether they have the necessary access, tools, and information to perform their tasks.
It's all about skill, governance, scope, and management.
And there it was, right before my eyes in those retrospective notes: a clear illustration of how people’s actions influence project outcomes. It could be a last-minute feature request from the client—poorly thought out but backed by a C-level executive, causing other work to be sidelined—or a client lacking the resources for user acceptance testing, leaving nothing feeling fully delivered.
Or mix-ups such as having front-end teams ready to deliver components that had not yet been designed by another agency. These are issues familiar to all of us who contribute to digital projects in one way or another!