With its ability to automate both high-volume transactional and cognitive tasks, Artificial Intelligence (AI) is an ideal match for enterprise-scale IT operations. Utilizing AI in this manner was the central theme of a breakout session presented by Stephen White, the Head of Intelligent Automation, Development and Technology at IT consulting firm GDT, during the recent Digital Workforce Summit (DWS).
GDT is a longtime IPsoft partner that has implemented dozens of IPcenter and Amelia deployments for clients. The company is now beginning to work with 1Desk, (including, according to White, an internal deployment that just recently went live). White said he views 1Desk as the combination of IPcenter’s autonomic framework with Amelia’s conversational interface – a means to “not just support [customers’] infrastructure, but also supporting their end users as well.”
While GDT intends to bring the 1Desk platform to its customers, White’s DWS session centered on the benefits and lessons learned from implementing IPcenter and Amelia. He noted that new companies beginning their AI journeys are often confused on how to properly measure success. He said a good yardstick is the ability to reinvest in other parts of the business thanks to new operational efficiencies derived from deploying automation and cognitive technologies.
“What we’re talking about is turning automation into the ability to free up the capacity [and] free up the dollars,” he said. “To shift those [resources] into those transformational efforts.”
White illustrated the ripple effects from AI-driven IT operations by discussing automation’s impact in his previous role with a major life insurance provider, where these technologies not only saved money but workers’ time. Before the provider added automation, “we had billions of dollars in IT spend and we didn’t have enough to spend on transformational efforts, because it was all being consumed by technology debt,” he said. “So automation began freeing that up for us… It allowed us to return that human capital back to [other IT groups] so they could reinvest that in what they wanted to work on.”
Better Data, Better AI
White said companies also should consider automation’s potential for increasing both the quantity and quality of data. “In the Machine Learning space, you have to have a lot of very good training data. Most companies just don’t have that. It’s a shortage of the volume. It’s incomplete, inconsistent [and] of low quality. Through automation, we can fix a lot of that,” he said. “First, companies come to us and say, ‘We have data problems.’ They don’t have data problems. They have process problems that are yielding incomplete and poor data quality. So if we fix the process, by automating that process, the residual of that is high-quality, consistent and complete data.”
“You don’t have to be an engineer, you can talk [to AI] like a normal person. Sometimes end users get frustrated with us because we make it so hard to provide [automated] services to them. Amelia is that bridge to unlock those automation capabilities.”
— Stephen White, Head of Intelligent Automation, Development and Technology at GDT
Conversational platforms like Amelia are uniquely suited for generating high-quality data by automatically translating user engagements into useful structured data. “About 92% now of the world’s data that is being created is unstructured. Most companies aren’t able to harness the value of that unstructured data at this point in time. That’s why we’re using Amelia,” White explained.
When Amelia handles a service desk issue, she automatically quantifies useful data (user ID, intent, attempted fixes, etc.). When it comes to generating insightful data that could be used for future refinements, White said user engagements with IT workers are inherently less efficient than those with a virtual agent like Amelia, who can automate data generation “so that we can actually mine it and use it to do the continuous service improvement activities.”
Of course, Amelia isn’t only a means to provide IT departments with better data – she provides better services to all users by lowering the barriers to interacting with automated systems. “You don’t have to be an engineer, you can talk like a normal person,” White said. “Sometimes end users get frustrated with us because we make it so hard to provide those services to them. Amelia is that bridge to unlock those automation capabilities.”
Better AI, Better Organizations
AI allows enterprises to use resources more judiciously and open access to services for all end users. However, it is also inherently useful to IT workers by freeing them from routine high-volume work. While many engineers are anxious regarding automation’s effect on their jobs, White said he addresses these concerns by asking, “‘What do you not want to do anymore? Give us that stuff,’” he explained. “Because invariably, [their answers] fall into the big bucket of high-volume, low-complexity stuff that anybody could do.”
In fact, White highlighted how those “high-volume, low-complexity” tasks are the best AI automation starting points. Once those tasks are mastered, companies can begin iterating up the complexity ladder. “Start freeing up that capacity. Then you can shift those resources to do those higher-value activities,” he said. “Also, as you start working your way up the complexity stack, look at automating problem management, for example. We’ve done that at a couple of organizations, where we go in and we’ll remediate the unhealthy assets out of the environment [and] actually drive the operational costs down, because we’re not spending hours and cycles fixing things that continually break.”
As the company begins the next leg of its automation journey with 1Desk, GDT intends to find new capabilities and efficiencies linking back-end services to front-end user engagements. As White concluded: “It’s just a beautiful thing to be able to step back, and look at your work and watch your digital labor force take off from there.”