Defining an Enterprise Automation Strategy

It’s critical to establish an automation strategy so that the company is aligned on the vision and can overcome scaling challenges.

Two popular frameworks used in information technology are Controlling Objectives for Information and Related Technology (COBIT) as well as Information Technology Infrastructure Library (ITIL).

COBIT is focused on controlling enterprise IT and establishing a roadmap for where it'll go. In other words, COBIT focuses on reducing risks and establishing a strategy.

ITIL is focused on continuous improvement and efficient operations — it focuses on best practices that improve outcomes for the business (Leinov, 2018).

You may think of ITIL as tactical and of COBIT as strategic. It’s possible your company is already using some of these libraries within specific departments (such as ITIL within IT support and COBIT within accounting). Which framework or combination of frameworks is used is something that your company has likely already decided upon, but it's good to know about alternatives and where you can borrow knowledge from.

The reality is that any framework for managing IT solution delivery is better than the ‘automate first, think about it later’ mentality.

Simply 'winging it' will result in an RPA program running out of steam, automating the wrong processes, or being tactical at the expense of not seeing the bigger picture.

ITIL and COBIT aren't the only frameworks in existence, and it's worth exploring other options if your company is in need of a best practices library. This article in particular will focus on the main takeaways from COBIT’s 'Manage Strategy objective' and how those insights apply to establishing an automation strategy.

Strategy Management Using COBIT 2019

According to Process Symphony, COBIT suggests that strategy management should involve:

To accomplish good strategy management, COBIT specifically recommends the following management practices:

APO0 2.01 - Understand enterprise context and direction:

Identifying what competitors are doing, where the industry is headed, and what regulations you’ll face in RPA are all important undertakings in this practice.

If you’re in the lending space and the competition is moving towards using AI to assess borrowers' risk, it’s important to recognize that trend and not over steer towards RPA automation. In the lending space example, you may want to explore both RPA and AI (known as intelligent automation) to remain on the cutting-edge.

If, on the other hand, you’re in the service industry with hundreds of employees working on IT demands spanning account management and troubleshooting, it'd be critical to identify how RPA can help and where the competition is heading.

Connecting with RPA software vendors or consultants through events, as well as reading their white papers, is a great way to become aware of the applications of automation in your industry. Remember that this information is biased with a skew towards the implementation, and payment, of RPA vendors or IT providers' services.

To remain impartial, you can 'mine’ the industry use-cases and do your due-diligence by connecting with other leaders in your line of work. Learn from their successes and also make note of their challenges. If the benefits you've heard of in RPA are tipping the scales in its favor, make sure you find counter arguments to develop a balanced view.

Set the context for what is and isn't in scope for RPA:

Letting IT service providers drive the narrative of your company's automation roadmap could result in a 'gold brick road' phenomenon. You don't want to stop looking at the road because you think 'it's taken care of' only to end up left dealing with a broken suspension after running into potholes...

Be sure to factor your organization’s tendency for innovation adoption as well. Is your organization an innovator, an early adopter, an early majority, a late majority, or a laggard?

If there’s only a handful of companies succeeding with RPA in your space, and they have budgets that are fifty times greater than yours, its possible that as an early comer you'll be better of waiting until at least twenty more companies in your industry announce success with RPA. Granted, the number twenty isn't set in stone - it's just an example, but such an approach not only gives you an opportunity to catch up with the technology, but it also gives the market a chance to learn your industry.

In a smaller company you'll also have less opportunities for automation (most of the time) and you'll have less margin for error, which means that it's better to let the ecosystem spend money digging new wells of productivity. After they've established where the 'oil is flowing' buying up real estate in the victinity to profit from their discovers is in tow.

Even though RPA has been gaining steam over the past half decade, hiring a tried-and-true RPA specialist continues to be a challenge. As time goes on, that should hopefully become less difficult. But why wait for the industry to catch up when you can train existing IT specialists on RPA within your company?

APO0 2.02 - Assess current capabilities, performance, and digital maturity of the enterprise:

If your company’s IT department is already using ITIL as a framework, then use whatever materials are available from that framework, or another management lens, to determine what the current digital capabilities are.

In this objective, you'll also assess the supporting IT capabilities of your company to determine whether it'll be cost-effective to establish RPA as an in-house function or if consulting is recommended.

One must recognize where their company is at from a digital maturity perspective.

All of these questions, and more, must be answered to determine where IT is at from a perspective of being able to deliver and oversee RPA implementations. Likewise, it's important to understand where supporting business teams are at in their digital maturity:

Assessing the company's maturity in digital data management is paramount to developing an enterprise RPA strategy.

For example, if hundreds of emails are sent in with scanned forms from customers, we'd say that the process is mostly analog — information has to be manually entered by an employee from the scanned forms. The delivery method of the form is digital, so the process is somewhat digitized, but not nearly enough to be ready to benefit from an RPA perspective. Sure, a third party IT service provider might tell you that they can get those forms read using Optical Character Recognition (if you pay them a fair amount of money and would like to purchase an OCR license from a vendor that integrates with your RPA solution), but that’s often not the best way to go it.

Consider for example that a form which is often typed out and printed by a customer would better serve the customer if available online so that they can fill it out on any of their devices and not have to worry about printing the form in advance (since not everyone has access to a printer). Furthermore, customers that are traveling wouldn’t be able to print out a form and would not be as satisfied having to wait a few days or weeks to get back home just to print a document to fill it out by hand.

APO0 2.03 Define target digital capabilities:

In the example above (where customers email forms), it'd be best for the organization to consider a customer relationship management system which can track a variety of requests. It'd also be great to intake information via digital forms, if the opportunity is there from an IT and enterprise prespective. Establishing such goals would allow the organization to plan for the upgrade of their systems, to stop relying on analog means, and to become fully digitized (with or without the requirement to implement any RPA).

Implementing digitization, at a level of optimal capability, should strive for a system which offers workflow automation out of the box. One can imagine in a customer request scenario that if a form is filled out with a specific product, then the task of reviewing and approving the customer’s details ought to go to a designated team as opposed to a general queue. This level of workflow automation, even though basic, would fulfill the concept of ‘digital transformation’ - the process would be digitally processed to some extent and not just 'digitized.'

Only after the organization has addressed its digital maturity and arrived at the most optimal capability of systems should it consider automation of remaining gaps. Otherwise, the company is at risk of automating bad processes such as those that rely on free-form email (which isn't scalable even with typed forms/OCR). The company is also at risk of funding something that'll never be profitable.

If it took 6 months to automate 10 processes using an email medium, and after 3 months the emailing system is replaced with a proper CRM, the bots would instantly become useless. Some companies can effectively predict the value of automation as being profitable, even if it'll be run for a temporary period (before a system migration a year or two out). There's no rulebook against building automation that has a known end-date from the very beginning, but it takes a serious consideration of the costs and benefitsto succeed in these edge-cases.

When the organization produces loss-makers, it becomes much more difficult to achieve success. This can lead to a frantic approach of automating things “faster” so that benefits can be realized before changes take place or automating as many things as possible, even if they provide small benefits.

To put this section simply, the target digital capabilities of an enterprise should not rest on RPA capabilities as a foundation. Digital capabilities refer to systems, processes, and integrations at the infrastructure level. Only after enterprise architects, process improvement teams, and system administrators have improved the system within their means should RPA be explored.

APO0 - 2.04 Conduct a gap analysis:

Analyze the current state as compared to the desired future state. Determine where gaps exist. This is dependent on the industry you’re in and the size of your company. There’s a lot of material on this topic, so we’ll let you dig deeper as needed.

APO0 - 2.05 Define the strategic plan and road map:

Without a plan, there’s no way to gauge whether you’ve succeeded or not. There’s also no way to get aligned with cross-functional stakeholders on how to achieve the vision and in what time frame to do so. At the very least, it's recommended to establish a six-month to a one-year plan for RPA proof of concept initiatives (as well as planning one to three years out after the first phase is completed).

For example, you might bring in consultants for 6 months to determine if RPA is practical at your company by creating five bots for specific processes that show the most potential for benefits and implementation success. During month three you’ll have to decide if you want to continue to support the bots that are being built (or have already been deployed), and you’ll likely be deciding whether this consultancy is the group you want to continue working with for the long-term.

The strategic plan may include:

1. How Automation Governance will be defined
2. Establishing where the RPA Center of Excellence (CoE) will be housed
3. The Automation Operating Model and cross-functional collaboration
4. Development resource allocation, training, or outsourcing
5. The cost-benefit model

APO0 - 2.06 Communicate the I&T strategy and direction:

After you’ve established the information and technology strategy, and received the necessary approvals from leadership, it’s now time to communicate the strategy. Change management practices, such as those from Prosci, may be helpful in meeting this objective.

Footnote:

The term 'Robotic Process Automation' is used due to its popularity in the industry. Process automation is a clearer term that'll ideally replace 'RPA.'

References:

Leinov, V. (2018, May 15). COBIT vs ITIL: a real-life example of what these two can do. ScienceSoft. https://www.scnsoft.com/blog/cobit-vs-itil

Process Symphony. (2019, June 2). Strategy Management APO02 (COBIT2019) – Process-Symphony – ITSM Knowledge Orchestrators. https://wiki.process-symphony.com.au/framework/lifecycle/process/strategy-management-apo02-cobit2019/