As leader of the Global IT Service Support organization for InterContinental Hotels Group (IHG®) – a world-renowned hotel company that includes well-known brands such as InterContinental, Crowne Plaza, Holiday Inn, Staybridge and Candlewood, in 100 countries – Scot Whigham leads a team that delivers IT support to 30,000 corporate employees and associates. His team makes sure that any IT issues are resolved quickly, so there’s no impact on IHG’s primary mission to deliver sterling service and hospitality to lodgers and guests. We spoke to Whigham about why he looked to AI and specifically to IPsoft’s Amelia as part of his service strategy, how his organization has adapted to working with AI, and his organization’s ongoing success with Amelia.
For those unfamiliar with Amelia, she is the world's most human and fully conversational AI, entirely automating human-to-human interactions and process execution. Like a human, Amelia communicates using natural language and can respond to customers’ emotional states carrying out depth of conversations far beyond what Siri, Alexa, Cortana or Watson can do. Unlike a human, she can hold thousands of conversations in parallel. Amelia is currently deployed only in the Enterprise.
What influenced you to consider AI initially? What drew you to the technology and then specifically to IPsoft?
What drew us to the technology was trying to solve the challenge that IHG gave to us which was expanding our capacity without expanding in the technical support field, providing customer [IHG Hotel and Corporate colleagues] service and providing key components of availability and being able to respond to our customers. We were given the challenge to meet those high demands from our customers for engagement and we couldn’t do it because we didn’t have the physical space and also meet our budget priorities and improve our quality. So it was “how do we expand without expanding?”
We started taking a look at chatbot technology and we quickly realized that it was not going to be adequate because the experience, the engagement with our customers was not personal, could not react in the moment, and the customer would have to train themselves to interact with the tool to get the information out of it. We would not be able to get significant information back from customers in order to continuously improve the product. From that we knew chatbot technology was not going to be part of the decision-making process.
We spoke to a number of different vendors. We selected IPsoft because of the tools and the team, and seeing the capability for our customers to have a conversational experience with Amelia. We saw the capability to have a real-life experience, very similar to how conversations take place between us [IT support] and our customers. Our end users are looking for support that they can engage with, just as they engage with our [hotel] guests.
Make sure [Amelia] is part of the overall strategy, and make sure that people are included in the strategy and explain how they’re going to work together with the technology. - Scot Whigham, Director Global IT Service Support, InterContinental Hotels Group
Now that you’ve used AI technology, what were your biggest surprises, and what were the biggest challenges?
Three things in particular. This is new technology and the necessary skill sets are few and far between. How do you scale up those skills? IPsoft has significant skills to deliver on the platform, but the key is also having the expertise on our internal business processes and how they would work in this platform. It’s critically important to understand the soft skills you need and understanding the business processes necessary to execute on different tasks. What was surprising was how poorly we documented the business process day in day out, and working with Amelia provided us a much greater understanding of all of our processes.
Second thing that was the level of training. We should have engaged in training before the [implementation] process even started, in order to get both sides [IHG and IPsoft] to speak the same language and terminology so we can effectively communicate in deploying this technology.
The third was how to let the technology’s true capabilities come forward, and not try to programmatically determine every little step in the process. We did not leverage the technology’s ability to learn and to map out conversations, to help build out the BPMs (Business Process Models), and to help us establish a baseline and a framework where we could quickly deliver. Due to the way we originally deployed, we slowed ourselves tremendously and we limited the amount of data we fed into Amelia. If we were to start all over again, we would change that up and try and feed all the data that we could to Amelia in terms of process, knowledge base and actual conversations that were taking place. We would allow Amelia to start mapping out as quickly as possible and allow her to build that baseline framework, that we could then hone those experiences we wanted our customers to have with Amelia.
How do you push culture change in an organization when implementing AI technology?
The cultural challenge that we had to meet with our tech organization was the fear that this was job-killing technology. That’s the fear we faced the most and what we’ve had to face in this technology. We didn’t want to overly promote this technology, saying that it was going to solve all challenges – it’s not a panacea or a silver bullet to any problems that you have. It should be part of your overall business strategy. It was critical to overcome the fears we had internally, so we flipped that on its head to demonstrate how staff would fit into the business strategy and how their jobs would fit into higher-level roles. What you see is our employee engagement scores remain extremely high within the support organization. Make sure this technology is part of the overall strategy, and make sure that people are included in the strategy and explain how they’re going to work together with the technology.
We’ve seen high satisfaction rates so far with those customers that have engaged with Amelia. - Scot Whigham, Director Global IT Service Support, InterContinental Hotels Group
What metrics do you have or think are important for measurement of success?
First and foremost, does Amelia help our customer? Is she successfully assisting the customer? We pay very close attention to her capability to execute on individual tasks and the different situations we train her on. For the initial business cases that we trained her on, she has had success rate in the 80-85 percent range. It’s the same experience if you measured individual agents, they’re ability to execute on tasks they’ve been trained on. One great advantage with Amelia is you train at one point, then for every interaction she has going forward, and because she can do transactions simultaneously, that means we have one point to change and she can execute on that change. It’s a critical metric that we use.
We’re deploying some other technology in order for us to understand the satisfaction of the experience with our customers, and how they feel about the technology.
We’ve seen high satisfaction rates so far with those customers that have engaged with Amelia. What we’ve been hearing from our customers is that “it’s about time” – they’ve been expecting this technology and our culture is embracing it.
Contact to learn more about Amelia and other cognitive tools.