The future of CX automation is here, and the question on everyone’s mind is where that leaves the people on your support team. In the latest episode of our Future-gazing AI series, our Director of Partnerships, Sylvain Mlodyszewski, and Kurt Pinoy, the CEO of Premium Plus, discuss the importance of strong business partnerships when it comes to implementing the most effective AI tools for your brand. They also explore the human factor in the future of CX automation, reflecting on how the most powerful automation solutions will be based on teamwork between AI and human agents.
Watch the full session here, or read the transcript below.
Sylvain Mlodyszewski: So hello everyone. Welcome to the series of the Future-gazing show of Ultimate. I'm Sylvain Mlodyszewski and I'm pleased to be here today with Kurt Pinoy from our partner Premium Plus. Kurt, welcome to the show. Do you want to introduce yourself very shortly?
Kurt Pinoy: Yeah, thank you. I'm Kurt, CEO of Premium Plus, one of the go-to market partners of Zendesk in EMEA. We are already 11 years there, so we have a lot of experience and are very happy to join you in this call.
SM: Thank you. And for today's session, we really want to take a broad approach and have a discussion about the developments of AI. It's on everybody's mind lately, and everyone is talking about it. There is news about it every day, and we're pleased to invite a few experts from different fields to get their opinion on the pulse of AI today. And in order to get the conversation going, I really wanted to ask you what short term changes do you see in the CX space specifically in the next six months for your industry?
KP: Yeah, we see, we see a lot of things happening as we speak. So we first had our own impression when ChatGPT entered and AI entered the arena, and it was a big wow effect. But I think in the first six months increasing automation of routine tasks is something we see done very well. Improving natural language processing and, let's say, sentiment analysis was really something very cool just in terms of better understanding what the customer wants and what their needs are. That's something that has been very well received in our company. As I said, at first was the wow effect. We built something ourselves in Zendesk to define, okay, what's possible with ChatGPT and AI. We were very impressed, and then we were just looking at how to optimize our internal processes, and overall interactions with customers. So we’re very excited about what's been happening in the last six months.
SM: Thank you. Also, I think we see that those late developments were really like a lot of quick, incremental additions to the same way that we are doing things, but trying to see how we could automate a few of these tasks. If you could now, for our audience, take out your crystal ball and now look – not in six months, but maybe in five to 10 years – and tell us where you think we are headed and what your predictions for five to 10 years would be? What could dramatically change in the way we do things in customer support?
KP: I think it would be the hyper-personalization of things. I think that's something that we’re seeing. AI can really really revolutionize the CX through improvements in predictive analysis based on user data. What really excites me is the seamless integrations across all channels, and ultimately transforming the way we see the customer journey as a whole.
SM: Thank you. That makes a lot of sense. And one question that comes to mind for you as the CEO of Premium Plus, your job is to provide services to companies and we have seen with this huge wave of AI, that there is a lot of rightful questioning about what this means for jobs, what this means for the workers working in these fields. So what's your take on AI taking jobs versus AI creating new opportunities?
KP: Well, in CX, let's say you have three different levels. I see self-service as one level where people expect things to be automated. Let's say if you have a bank app, you don't want to call or contact somebody to just know how much money you have in your bank account. But if you can't pay anymore, then you need to make it personal and be able to speak to an agent. So the human interaction will also be very important, meaning that for self-service, you can automate a lot of things. But a good mix between both will be very efficient. In the agent workspace, if you look at sentiment or AI making a summary of a long text or providing short reply text that can just empower agents to become better.
"Nobody likes to do repetitive tasks. So if you can make sure those are automated, and make sure that the really tricky conversations go to your agents, that will just empower them and it'll also make their role more attractive. "
On the other hand, the third level is data analysis. If you have a good self-service and a good agent workspace, you need to know what happens in the company. And with AI-powered analytics, it can give you insights on what's really happening, and go back to the process and change it if needed.
So I see a lot of opportunities with tailor-made solutions for us as a service provider. We even build our own ChatGPT apps on the Zendesk platform. So I see opportunities, and new opportunities and conversations in data analysis. So it's a good mix.
"I think AI will never take everything over. It's a good mix between the repetitive tasks being automated, and delivering a personal experience whenever it is needed because that will make ambassadors of your clients. That's the goal of our existence here."
SM: Great, right. Yeah, absolutely. And also, speaking of empowering the agent, empowering people more than trying to replace them, I think is the right approach. Can you share, from your perspective, tips and tricks for how you and the team at Premium Plus have adopted some of these new automation tools inside of your daily operations?
KP: Yeah, it was not orchestrated, but at a certain moment I was just looking in our own support agent workspace, and I saw an app that our people were building to get a summary to reply. My team was just experimenting with whatever the possibilities are. I think as one of the leading go-to-market partners, I really am proud to have such a team that is always looking to innovate and rethink the things we do. So I just saw what they were doing in our own environment and I said, why the hell are we not making new apps out of that? And we made one Summary App that just makes a summary of intents, and then the Reply App was also a cool app that we launched.
"So I think everything starts with listening to our clients, and then seeing whatever opportunities there are. Seeing with you guys or with other tech partners what the options are and then making the best solution possible for our clients. That's our goal."
Listen and then make suggestions. I must say, our clients are really looking forward to whatever AI brings to the table, but they also see some risks. But in our case, all departments are using it now. Not only using it and then just not making a control out of it, but just using it in a good way and then personalizing it whenever needed. That's also my advice, but I’m very happy to see what our team has been developing in the last six months.
SM: Great to hear that. Actually I'd like to follow up on a topic. Now, with the huge hype around AI and ChatGPT, a lot of clients are, I imagine, a bit lost in this ocean of information. And so I assume it's your place also as a service provider to help your clients navigate these waters and really identify the right solutions. So is that something where your clients come to you and say, ‘Hey, I heard those AI buzzwords, can you help me navigate this and how I can leverage these new technologies?
KP: Yeah, 100%. I think the tool is one way to reach your end goal, but the guide is, in our belief, the most important one that will bring the best practices not only with Zendesk, but let's say, they are looking for a phone solution. If you go to the marketplace of Zendesk, you will find 200 solutions and they're lost. So we try to be good experts in Zendesk and collaborate with tech partners where they have added value on top of the features of Zendesk. We guide them or we do it ourselves, but the most important thing is that we listen and then try to guide them in the community to the best partner for a specific use case and connect people. I also don't see other partners as competitors. I think we all have our place, we all have our specialties, and we should collaborate and instead of competing against each other, that's how I see business. That's what we have been doing for 11 years. We would not be here if we had not collaborated with all our tech partners.
SM: I think the same way. I work in partnerships, and I'm really glad that you and the team at Premium Plus have also made the choice of collaborating with Ultimate as a tech partner and also trusting our technology to be the right fit for your customers.
Bringing it back to AI in the broad sense, one thing you mentioned in one of your last answers was you talked about the risks inherent to AI, and there are risks. Can you maybe develop a bit on this point from your own experience and what you hear from your customers?
KP: I think you should really define very well the scope of AI, where the resources are to guide customers with content because you can't open it for everything. The resource needs to be very well controlled in terms of where you pull all the data so that you don't allow it to use fake news or whatever that exists on the web. So the ability to make a really defined scope is something I see as a pitfall, but we are working on that — I think that's number one.
And number two, I think, let's call it human oversight of checking if the answers are really correct,
"I think especially when it comes to sensitive customer interactions, it should be about building trust in AI by creating a good mix between what can be automated and what should be handled through personal contact with an agent."
That's where I think it’s a good mix, and everything being AI would be a bad idea I think. But finding that good mix to build trust, and upscale customer satisfaction because if people will love it, if it's personalized, if it's to the point, and if they can skip it and go to personal contact, then you've got it I think. But that's what I see as risks.
SM: That makes total sense, and I think we see eye to eye on this. Obviously with our platform, we released UltimateGPT, which enables companies to have in a few seconds a new chatbot that is limited to the knowledge base and help center that they would provide. And we have machine learning engineers and a research and development (R&D) team that has worked extensively to limit what we call ‘hallucinations.’ So basically where the AI would generate wrong facts. So there is definitely a lot of work that needs to be invested upstream in order to control and create a solution that looks super standardized. At the same time, where we have invested in the backend, those R&D efforts to make sure that from the get-go, you can just plug it in and it just works. And I think that's a very important aspect.
So Kurt, thanks a lot for today, and for this session. Just before we part ways, do you have any final words, or anything else that you'd like to share with us today?
KP: I think what I wanted to share is that,
"We should embrace AI and then combine the strengths of AI with human empathy and creativity. And that will lead to a really, truly great customer experience."
So that's the main message I want to give. There are risks. There are opportunities. A lot of things are happening, but I see it as an opportunity, and I’m looking forward to the future.
SM: Thanks a lot Kurt for today and thanks for your time.
KP: Thank you. See you. Bye-Bye.