putilov_denis - stock.adobe.com


5 steps to crafting a successful automation strategy

IT leaders who bypass strategy and go straight to selecting automation tools are courting failure. Here's how to move toward enterprise automation successfully.

Putting the cart before the horse is a bad idea. Another bad idea: rushing to enterprise automation without a solid strategy.

Automation is one of the most hyped technology trends. But, as with any technology implementation, automation tools won't be successful unless they have the right strategic underpinning.

With that in mind, here are five steps to creating a successful automation strategy.

1. Start with business strategy

The urge to cut to the chase and start evaluating products is widespread and persistent. Too often, the result is another round of: "Here are the new tools. What was the problem?"

Focusing on automation tools before deciding on automation strategy can forestall exploration of other technologies that might be more helpful in achieving business goals, such as digital transformation. Some of these might even already be a part of the organization's technology environment. In addition, focusing too soon on automation technology -- or any particular technology, for that matter -- might mask the need for nontechnological changes, such as process or organizational changes. Focusing on technology prematurely also creates the danger of someone focusing too much on a specific technology or vendor at the expense of competing ones in the same technology area.

To buy the right tools, IT needs first to understand the problems it needs to solve. To assume the organization needs an automation strategy is a recipe for failure. Decision-makers need to skip past automation hype and ask: Does the organization actually need an automation strategy, and if so, why?

In other words, CIOs and other IT decision-makers need to start with the organizational strategy.

Part of that is determining what the business is trying to accomplish this year and by what means.

Organizational strategy determines whether automation is helpful in accomplishing those goals and to what extent. For example, "increasing profits by responsibly lowering costs" may be central to the business's plans for the year, or the goal might be "increasing revenue by sustainably serving more customers." Automation can play a key role in either pursuit. On the other hand, automation would serve a much narrower role in supporting business strategy if the main goals for the next year or two are focused on acquisitions. IT would probably still want to pursue automation tactically to increase the efficiency and improve the lives of IT employees, but a broad enterprise automation strategy would probably not be needed and might be a distraction.

A new strategy -- including an automation strategy -- should embody the key principles around which IT is organizing its efforts.

2. Identify overarching technology drivers

The degree to which automation should be a priority for IT is directly dependent on the overarching IT technology vision.

This is inside baseball for IT but important nonetheless. Any technology strategy should not just advance the goals of the business, but should do it in a way that supports IT leadership's long-term vision for IT. A new strategy -- including an automation strategy -- should embody the key principles around which IT is organizing its efforts. If IT is focusing on implementing zero trust in networks, data access and application use, then any new strategy should embody those principles.

Later on, in evaluating possible ways of implementing a strategy, one of the criteria on which to judge is the degree to which a specific tool, process, team or staff position advances one of these principles. Moreover, it should be possible to map from the business goals to the IT principles. The latter should support the former.

3. Determine architecture

IT leadership, architects and senior technologists next need to define the environment required to achieve the desired goals within the constraints, guideposts or guardrails of the technology principles.

This ideal future-state architecture might involve many technologies and processes and describe an environment reliant on staff from a variety of teams. It should highlight, at a high level, where critical information flows.

If there are constraints on specific components -- for example, that they must be in the cloud or that they cannot be allowed to be -- the architecture should reflect that, too.

4. Mind the gaps; build the roadmaps

Next, IT needs to evaluate what is already in place against that desired automation end state. This includes identifying which of the following areas they need to address:

  • new technology they need to acquire or existing tools they need to use differently;
  • new processes they need to define or where old ones need to be amended;
  • new teams or where existing ones need to be reconfigured or repurposed; and
  • new necessary skills that staff need, whether acquired through learning or through hiring.

With that set of differences in hand, IT can develop the roadmap for getting from the as-is to the to-be most effectively. In effect, this is the stage where IT defines automation strategy.

5. Implement the strategy

Where the roadmap requires acquiring new technologies -- which, in this case, could mean any of the automation technologies available, ranging from tools to support more mature, enterprise-grade scripting all the way up to low-code tools or even AI-powered robotic process automation tools -- IT has decide what to buy. It should follow a consistent evaluation and selection process in doing so.

The strategy likely also requires shifts in processes, team structures, staffing and staff duties. IT needs to work within existing HR and organizational frameworks for accomplishing these and should treat them as equal in importance to getting the tools. Staff training -- or time for skills acquisition -- is crucial where new tools are coming in. Just layering on another tool without making sure it can and will be used as intended often results in simply adding another layer of capabilities to the stock of shelfware and a failed automation initiative.

Next Steps

AI and automation take center stage at AnsibleFest 2023

Dig Deeper on Digital transformation

Cloud Computing
Mobile Computing
Data Center
and ESG