Why enterprises need an ‘evolutionary approach’ to IT automation

IT automation has been a topic of growing conversation amongst organisations looking to boost their competitiveness. In today’s dynamic business environment, enterprises need to be increasingly agile to ensure they have the speed required to meet changing market expectations. In technology terms, this comes down to ensuring the IT landscape of a company is both synchronised and optimised to provide the company with bandwidth, scale and muscle to re-calibrate quickly when required. The cost savings and efficiency benefits are of course an added bonus.

Despite these apparent merits, most automation projects have shown a low success rate. In a recent industry survey by eDSG of CIOs at Global 1000 organisations, 72 per cent indicated that they struggle with IT Automation Silos. A further 93 per cent of the respondents indicated that their organisations have little to no technology solutions that enable them to truly monitor, analyse and automate enterprise-wide business processes.

These results are not especially surprising, because in an enterprise IT environment where standardisation is still a distant dream, trying to apply futuristic automation without having the right foundations often leads to failure. Also, most of these programs are resource-reduction rather than resource-evolution focused, which leads to cultural friction and knee-jerk management decisions.

Three cornerstones of a successful automation strategy

With this in mind, a successful automation strategy requires three cornerstones:

1. It is evolutionary

Automation isn’t a ‘set-it and forget-it’ process; it must be realistic and gradual. A good automation strategy is evolutionary in so far that it aspires to progress block-by-block, enabling people and platforms to move up the value chain by first automating standard, repetitive tasks, providing a launch pad for the larger program.

2. It is business-aligned

First off, the ‘automate all things’ thinking needs to be replaced with an understanding of the actual goals and benefits of automation. Trying to automate too much is the only thing worse than not automating at all. The goal of IT automation is to fulfill the outcome of each value chain that impacts upon consumers and businesses. A successful automation approach should allow IT teams to both Request to Fulfill (R2F) and Detect to Correct (D2C). Automation programs should also be closely aligned to the business process maturity of an organisation, which is classified into four stages - ‘Reactive’, ‘Active’, ‘Dynamic’ and ‘Predictive’. A very different set of levers has to be plugged behind each of these stages. For instance, to achieve ‘Predictive Automation’, the right automation levers would be machine learning, operations analytics and cognitive expert systems and would require domain focus on software defined infrastructure, enterprise platform as a service and hybrid cloud. Similarly, ‘Dynamic Automation’ could be achieved with automation levers such as enterprise self service, multi-supplier integration, cloud lifecycle management, request automation, DevOps environment and cloud SSO.

3. It starts beneath the surface

An ideal automation platform operates as three layers (L-2 to L0) below the usual IT surface and automates at machine, platform and application level by creating preventive and self-healing systems and software, utilising techniques like machine learning and creating self-service capabilities with cognitive AI agents embedded into the systems. This ‘Beneath the surface’ optimisation ensures that the human talent above the surface can focus on more intellect and analysis-driven tasks.

Implementation best practice

To ensure a proactive and evolutionary automation strategy, it should be delivered in three waves. The first wave should involve using proven business service management methodology, lean IT frameworks, IT service management and levers such as unified monitoring, event automation and patch management. The second wave should then aim at identifying repetitive tasks, high-volume activities and complex undertakings with the view to automating them. The final wave should identify business driven areas, which need dynamic and agile automation, and look to apply modern methodologies such as machine and cognitive learning to the levers identified in the first wave.

The final word

Automation helps enterprises to stay in line with market and compliance expectations while achieving agility, accuracy, and competitive advantages. All it needs is an incremental approach and the right set of rules for it to benefit all the stakeholders.

C Vijay Kumar, President - Infrastructure Services, HCL Technologies

Image Credit: Shutterstock/Sergey Nivens