Image of hand pointing to the word Automation

Source: Adobe Stock

A study by Lopez Research found that 80% of organizations are looking to embrace and expand automation efforts in 2022. Today, market leaders are also using artificial intelligence to automate processes, find patterns in data and model outcomes of various actions.

There are many definitions of automation; not all are considered artificial intelligence. Automation describes a wide range of technologies that reduce human intervention in processes. These solutions range from more basic automation, such as robotic process automation, that uses scripts to emulate human processes, such as extracting data, filling in forms, and moving files. Many companies started with robotic process automation, but the automation field is much more comprehensive than this. At the higher end of the practice, automation extends to using artificial intelligence to understand and react to conditions with minimal human intervention.

I recently had several conversations with NTT DATA about intelligent automation, including an interview for the AI with Maribel Lopez (AI with ML) podcast that you can listen to here. A key takeaway from my conversations with the company is that there’s no one-size-fits-all approach to adopting automation. Organizations can use automation to help alleviate the impact of labor shortages by reducing human intervention for repetitive tasks. Still, in other cases, you may want more advanced AI-enhanced automation that streamlines alerts and suggests subsequent actions. For example, AI-enhanced network management may assess a variety of alerts, define which alerts are the most serious, and suggest network configuration changes to remediate an issue. A company can even set these solutions to make changes, if desired, automatically. My other takeaways are as follows:

1. Focus on the problem, not the technologies.
NTT DATA shared that a company needs to define the problem it’s trying to solve before the firm can successfully design a technology plan. It said, “We don’t define automation as technology. Automation is a set of tools and techniques to solve a business problem. We’re not looking for Robotic Process Automation (RPA), chatbot, machine learning, or IoT opportunities. (Instead), let’s first try and understand what we’re trying to solve and figure out if automation is the right solution.”

2. Map automation to business key performance indicators.
There are many areas where automation can assist the business. How do you decide where to start? NTT DATA spoke of mapping projects back to one of the three big levers of creating business value within a company. Does it help you grow revenue, optimize costs, or reduce organizational stress? These are the three major drivers of enterprise transformation for every type of company.

As I work with IT leaders on defining areas where automation can significantly move the needle on business value, I’ve discovered these projects often require a more significant process transformation. Most companies struggle with the process reinvention aspect of automation but automating an inefficient process won’t yield the best results. A company should streamline the workflow first, then automate necessary but routine functions. For example, credit checks within a mortgage application process are easy to automate, but if there are five unnecessary steps before the credit check, automating a single aspect of the process will have minimal impact. The company can still achieve the benefits of automation by focusing on several “quick wins” where they can rapidly demonstrate the value of automation

3. What gets measured gets improved.
Like any other project, it’s important to define metrics and procedures to measure success at the outset. In my experience, most organizations fail to specify if they are measuring hard or soft goals. NTT DATA calls this measuring return on value versus return on investment because some returns are qualitative versus quantitative. At some point, management will ask you to quantify the value automation has created for the organization. At times, a business can quantify the value of automation in dollars. At other times, the value gets measured as process acceleration or what tasks no longer need to be performed.

For example, employee experience and retention may improve due to automation minimizing manual labor. However, these types of metrics are difficult to equate to one item. NTT DATA shared that companies need to monitor automation outcomes to ensure everything works as planned and continues to perform well over time. Lopez Research reports that creating a lifecycle management approach to validating and refining automation is particularly important in AI-based automation, where machine learning can modify processes in unintended ways. The takeaway? Don’t forget data and automation governance, monitoring, and management.

Automation in the real world
All of this sounds great but is it practical? Over the course of several meetings, the company shared several examples of automation, including a case study they had published on the company’s work with Integra Lifesciences. This case study highlights three different ways automation was used within an organization to achieve both quantifiable benefits as well as employee experience improvements. For example, Integra LifeSciences used NTT Data’s Nucleus AI platform to speed up Oracle ERP testing by 98%. It also achieved a 50% faster increase in processing Oracle ERP user access requests through digitizing paper-based forms. 

As part of process transformation efforts, Integra also worked with NTT DATA’s Digital Experience designers to implement social listening technologies to actively track, gather and analyze data from the social media and online platforms favored by doctors and other users. Meanwhile, round-the-clock access to an Intelligent Assistant on Microsoft Teams, powered by Nucleus, allows employees to resolve issues at a time that works for them.

Things to remember
As your business upgrades its technology portfolio, a certain amount of AI and automation will be built into your business’s software and cloud computing solutions. There are several questions you should ask as you progress in your journey. What does your technology team need to create versus what comes inherently in the product? Do you have the internal skill set to develop automation, or do you need to select a partner? If you choose a development partner, does the vendor support a wide range of solutions so you can choose what’s right for you? These are just a few questions you’ll have to answer to make the most of your automation strategy.

In the post-COVID era, organizations have experienced at least the first wave of digital transformation. Now companies are spending more time creating technology strategies that enable digital acceleration. Automation is a fundamental component of this strategy. Effectively, automation will help you evolve from accelerating simple repetitive tasks to creating intelligent systems.

Other areas for you to consider
Machine learning and automation are also heavily used in security. If you’d like to hear more about ML and security, please check out this podcast with Lacework. As you look to pursue automation and insight through AI, you must build ethical AI. You can find an article I wrote on ethical AI here.

I look forward to sharing more with you on automation in the future.