Progress and digital technology have reached an interesting tipping point. According to Samuel Arbesman, in his book Overcomplicated, today’s computer systems are intricate to the point that we can’t fully understand or predict their behaviours. This doesn’t mean that we’ve lost control, but instead, it highlights that we should approach and manage technology in new ways.

Such complexity can appear in surprising places. For example, software as a service (SaaS) is often promoted for its simplicity. Pay for access, tweak it to fit with your people and policies, and stop worrying about the machines. It costs less, deploys quickly, scales brilliantly and returns capacity to IT staff. Simple!

But that isn’t the whole picture. SaaS still needs to fit into a context, or else you end up with shadow IT – the practice of unsanctioned services crashing the party. Shadow IT is appealing because it’s easy. Need storage for a team? Open a Dropbox account. Want to share some useful information quickly? WhatsApp is so convenient for that. Expect easy access to send an e-mail? Gmail is right there. Yet every such action risks damaging processes, pushes data away from the company mothership and creates security and compliance risks.

Shadow IT is an extreme example. But SaaS can grow from simple service to complicated management even in an official capacity, says Neil Jackson, Business Unit Manager for Axiz’s Technology Services: “When SaaS integrates with a business environment, the requirements become more complicated. And it’s an issue that grows when you scale up. This is something we see often: people think SaaS is simple because it looks simple in small deployments. But add users and your other requirements grow exponentially, and they also become more interlinked with other areas of IT, such as security and user personas. These responsibilities can eventually consume whatever resources you thought you opened by using SaaS.”

Managing today’s complexity

Fortunately, the problem of SaaS complexity isn’t unique. Current digital technologies are generally distinguishable for their complexity. Even though past technologies were not simplistic, today’s IT estates rely on a cat’s cradle of integrations and collaborations.

Technology management’s paradigm has shifted from a hardware-centric to a software-first mindset, and with that came many more ways to create solutions. Systems are more interdependent and modular, and like a stone in a rock pool, a minor issue can eventually grind a gaping hole into fragile technology ecosystems.

The same paradigm also provides a useful solution. Says Jackson: “Managed services have become the dominant way to deal with complex technology investments. Using software, you can get incredible visibility of different systems, even if they are somewhere else, such as a data centre or a third-party cloud provider. Managed services combine data with specialised teams to monitor, report on and manage complicated systems.”

As with SaaS, the customers of managed services don’t have to invest significant resources into technology. Instead, they use a managed service provider that takes care of orchestration and management responsibilities, aligning those systems with customer needs.

In his book, Arbesman states that the solution is to respect a system’s complexity and continuously study it to understand its behaviour. This approach is essentially what managed services do. They study the feedback from different IT components and processes, then respond accordingly to keep the system at their best. They ensure that complexity doesn’t rob a company of the value it gets out of technology.

SaaS, managed services and untangling complexity

Managed services are typically associated with infrastructure and resiliency management. Yet, as noted above, SaaS has all the potential to get out of hand. This may not be a big problem if a company uses a pure-service play among a handful of employees. Then, risks such as policy enforcement and data management aren’t tremendous considerations. But that situation is the exception. The rule is that SaaS will join and interact with a complex environment.

“Above a certain scale, SaaS makes more and more sense when it integrates with your other technology assets,” Jackson explains. “You rarely use SaaS without some type of networking and security component, and you’ll need things such as data access and process automation to get the most out of many SaaS products. You also have to keep an eye on not creating silos. SaaS is great, until you get stuck with a service you no longer want. At a certain point, SaaS becomes a key part of the areas that managed services look after.”

SaaS isn’t simple. When we forget that, we end up with digital white elephants or shadow IT. Don’t apply the traditional ways of managing technology. Those rules no longer work. The new way to tackle overcomplexity is through dedication, awareness and focused interventions – sometimes before an issue has time to take hold. The managed service way.

By James Francis for Axiz