Today, interactions matter greater than ever. In line with data compiled by NICE, once a consumer makes a buying decision for a services or products, 80% of their decision to maintain doing business with that brand hinges on the standard of their customer support experience, in response to NICE research. Enter AI.

“I feel AI is becoming a very integral a part of every business today since it is finding that sweet spot in allowing businesses to grow while finding key efficiencies to administer that bottom line and really do this at scale,” says vp of product marketing at NICE, Andy Traba.

When many consider AI and customer experiences, chatbots that give customers more headaches than help often come to mind. Nevertheless, emerging AI use cases are enabling greater efficiencies than ever. From sentiment evaluation to co-pilots to integration throughout the complete customer journey, the evolving era of AI is reducing friction and constructing higher relationships between enterprises and each their employees and customers.

“After we take into consideration bolstering AI capabilities, it’s really about getting the precise data to coach my models on in order that they’ve those best outcomes.”

Deploying any technology requires a fragile balance between delivering quality solutions without compromising the underside line. AI integration offers investment returns by scaling customer and worker capabilities, automating tedious and redundant tasks, and offering consistent experiences based on collected and specialized data.

“I feel as you are hopefully venturing into leveraging AI more to enhance your online business, the important thing advice I would offer is simply to deal with those crystal clear high-probability use cases and get those early wins after which reinvest back into the business,” says Traba.

While artificial intelligence has increasingly grabbed headlines in recent times, augmented intelligence—where AI tools are used to reinforce human capabilities fairly than automate them—is worthy of comparable buzz for its potential in the client experience space, says Traba.

Currently, the client experience landscape is extremely reactive. Looking ahead, Traba foresees a shift to proactive and predictive customer experiences that mix each AI and augmented intelligence. Say a customer’s device is reaching its end-of-life state. Reasonably than the client reaching out to a chatbot or contact center, AI tools would flag the device’s issue early and direct the client to a live chat with a representative, offering each the efficiency of automation and personalized help from a human representative.

“Where I see the long run evolving by way of customer experiences, is being rather more proactive with the convergence of knowledge, these advancements of technology, and definitely generative AI,” says Traba.

Full Transcript

From MIT Technology Review, I’m Laurel Ruma and that is Business Lab, the show that helps business leaders make sense of latest technologies coming out of the lab and into the marketplace.

Our topic is constructing higher customer and worker experiences with artificial intelligence. Integrating data and AI solutions into on a regular basis business may also help provide insights, create efficiencies, and release time for workers to work on more complicated issues. And all of this builds a greater experience for patrons.

Two words for you: augmented intelligence.

My guest is Andy Traba, vp of product marketing at NICE.

This podcast is produced in partnership with NICE.

Welcome Andy.

Hi Laurel. Thanks for having me.

:Well, thanks for being here. So to set some context, could you describe the present state of AI inside customer experience? Common use cases that come to mind are chatbots, but what are another applications for AI on this space?

:Thanks. I feel it’s a terrific query to start, and I feel at the beginning, the usage of AI is growing in every single place. Definitely, we had this big boom last yr where everybody began talking about AI because of ChatGPT and quite a lot of the advancements with generative AI, and we’re actually seeing rather a lot more doing now, moving beyond just talking. So just growing a use case of attempting to apply AI in every single place to enhance experiences. Certainly one of the more popular ones, and this technology has been around for a while, is sentiment evaluation. So as an alternative of just proactively surveying customers to ask how are they doing, what was their experience like, using AI models to investigate the conversations they’re having with brands and routinely determine that. And it is also a very good use case, I feel, to emphasise the importance of knowledge that goes into the training of AI models.

As you consider sentiment evaluation, you should train those models based on the actual customer experience conversations, possibly past records and even surveys. What you should avoid is training a sentiment model possibly based on movie reviews or Amazon reviews, something that is probably not well connected. So actually sentiment evaluation is a very fashionable use case that goes beyond just chatbots.

Two other ones I’ll bring up are co-pilots. We have seen, actually, quite a lot of recent news with the launch of Microsoft Copilot and other forms for copilots inside the contact center and definitely helping customer support agents. It is a very fashionable use case that we see. The rationale driving that demand is the varieties of conversations which can be attending to agents today are rather more complex. AI has done a very good job of taking away the simple stuff. We now not need to call right into a contact center to reset our passwords, so what’s left over for the agents is rather more difficult varieties of interactions.So with the ability to assist them in real time with prompts and guidance and recommending knowledge articles to make their job easier and more practical is de facto popular.

After which the third and final one just on this query is the really type of rise of AI-driven journeys. Many, a few years ago, you and I’d call right into a contact center, and the one channel we could use was voice. Today, those channels have exploded. There’s social media, there’s messaging, there’s voice, there’s AI assistance that we will chat with. So with the ability to orchestrate or navigate a customer effectively through that journey and recommend the following best motion or the following best channel for them to cut back that complexity is de facto in demand as well. And the way can I even get to some extent where I can proactively engage with them on the channel of their selection on the time of day that we’re prone to get a response is actually an area that we see AI playing a crucial role today, but much more so in the long run those three really sentiment evaluation, the rise of co-pilots after which using AI across the complete customer journey.

In order AI becomes more popular across enterprises and across industries, why is integrating AI and customer experience then so crucial for today’s business landscape?

I feel it is so crucial today since it’s finding this sweet spot by way of business decision-making. When we predict of business decision-making, we are sometimes challenged with, am I going to deal with revenue or cost cutting? Am I going to deal with constructing recent products or perfecting my existing products? And barely has there been a technology that has allowed a business to realize all of those directly. But we’re seeing that today with AI finding a sweet spot where I can improve revenue and keep customers completely satisfied and renewing and even gain recent ones without having to spend additional money. I could even do this in a more efficient way with AI. Inside AI, I can take a really modern approach and produce recent products that my customers demand and save money and time through efficiencies in making my current products higher. I feel AI is becoming a very integral a part of every business today since it is finding that sweet spot in allowing businesses to grow while finding key efficiencies to administer that bottom line and really do this at scale.

And speaking of those efficiencies, worker experience lays that foundation for the client. But based in your time at NICE and inside business operations, how does worker experience affect the general experience then for patrons?

I feel what we have seen at NICE is de facto that customer experience and worker experience are hand in glove. They’re one and the identical. They’ve tremendous correlation between one another. Some examples, just to offer some anecdotes, and that is customer experience really happening in every single place. In the event you go right into a automobile dealership for a Tesla or a BMW, a high-end product, but you might be interacting with a salesman who’s a bit of pushy or possibly just having a nasty day, it’ll deteriorate the general customer experience, in order that bad worker experience causes a negative effect. Same thing if you happen to go to your favorite local restaurant, but you possibly have a brand new server who’s probably not well-trained or remains to be determining the menu and the logistics that is going to have a negative spillover effect. After which even on the flip side of that, you may see worker experience having a positive effect on their overall customer experience.

If employees are engaged and so they have the precise information and the precise tools, they will turn a negative right into a positive. Consider airlines, a really commoditized industry without delay, but when you’ve an issue together with your flight and it got canceled and you’ve a critical moment of need, that worker from that airline could really turn that have around by finding a brand new flight, booking you, ensuring that you just are in your trip and meeting your destination on time or without little or no delay. So I feel after we take into consideration experiences at large and the worker and the client outcomes are very much tied together, we have done research here at NICE on this exact topic, and what we found was once a consumer makes a buying decision for a selected services or products, after that time, 80% of that consumer’s decision to proceed doing business with that brand relies on the standard of their interactions.

So how those conversations play out, plays a really, very necessary a part of whether or not they may proceed doing business with that brand. Today, interactions matter greater than ever. To conclude on this query, one among my favorite quotes, customer experience today is not just a part of the business, it’s the business. And I feel employees play a very necessary front role in achieving that.

That actually is sensible. 80% is a big number, and I feel of that in my very own experiences, but could you explain the difference between artificial intelligence and augmented intelligence and likewise how they overlap?

Yeah, it’s a terrific query. I feel today artificial intelligence is actually capturing all of the thrill, but what I feel is just as buzzworthy is augmented intelligence. So let’s start by defining the 2. So artificial intelligence refers to machines mimicking human cognition. And after we take into consideration customer experience, there’s really no higher example of that than chatbots or virtual assistants. Technology that permits you to interact with the brand 365 24/7 at any time that you just need, and it’s mimicking the conversations that you just would normally have with a live human customer support representative. Augmented intelligence alternatively, is de facto about AI enhancing human capabilities, increasing the cognitive load of a person, allowing them to do more with less, saving them time. I feel within the domain of customer experience, co-pilots have gotten a very fashionable example here. How can co-pilots make recommendations, generate responses, automate quite a lot of the mundane tasks that humans just do not like to do and admittedly aren’t good at?

So I feel there’s a transparent distinction then between artificial intelligence, really those machines taking over the human capabilities 100% versus augmented, not replacing humans, but lifting them up, allowing them to do more. And where there’s overlap, and I feel we’ll see this trend really start accelerating within the years to are available in customer experiences is the mix between those two as we’re interacting with a brand. And what I mean by that’s possibly starting out by having a conversation with an intelligent virtual agent, a chatbot, after which seamlessly mixing right into a human live customer representative to play a specialized role. So possibly as I’m researching a brand new product to purchase akin to a cellphone online, I can have the option to ask the chatbot some questions and it’s referring to its knowledge base and its past interactions to reply those. But when it is time to ask a really specific query, I may be elevated to a customer support representative for that brand, just might decide to say, “Hey, when it is time to buy, I need to make sure you’re talking to a live individual.” So I feel there’s going to be a mix or a continuum, if you happen to will, of a majority of these interactions you’ve. And I feel we’ll get to some extent where very soon we may not even know is it a human on the opposite end of that digital interaction or simply a machine chatting forwards and backwards? But I feel those two concepts, artificial intelligence and augmented intelligence are actually here to remain and driving improvements in customer experience at scale with brands.

Well, there’s the client journey, but then there’s also the AI journey, and most of those journeys start with data. So internally, what’s the strategy of bolstering AI capabilities by way of data, and the way does data play a task in enhancing each worker and customer experiences?

I feel in today’s age, it’s normal understanding really that AI is just nearly as good as the information it’s trained on. Quick anecdote, if I’m an AI engineer and I’m attempting to predict what movies people will watch, so I can drive engagement into my movie app, I’ll want data. What movies have people watched up to now and what did they like? Similarly in customer experience, if I’m attempting to predict one of the best final result of that interaction, I need CX data. I need to know what’s gone well up to now on these interactions, what’s gone poorly or unsuitable? I don’t need data that is just available on the general public web. I want specialized CX data for my AI models. After we take into consideration bolstering AI capabilities, it’s really about getting the precise data to coach my models on in order that they’ve those best outcomes.

And going back to the instance I brought in around sentiment, I feel that reinforces the necessity to be sure that after we’re training AI models for customer experience, it’s done off of wealthy CX datasets and not only publicly available information like a few of the more popular large language models are using.

And I take into consideration how data plays a task in enhancing worker and customer experiences. There’s a method that is necessary to derive recent information or derive recent data from those unstructured data sets that usually these contact centers and experience centers have. So after we take into consideration a conversation, it’s extremely open-ended, right? It could go some ways. It is just not often predictable and it’s extremely hard to know it on the surface where AI and advanced machine learning techniques may also help though is deriving recent information from those conversations akin to what was the buyer’s sentiment level originally of the conversation versus the tip. What actions did the agent take that either drove positive trends in that sentiment or negative trends? How did all of those elements play out? And in a short time you may go from taking large unstructured data sets that may not have quite a lot of information or signals in them to very large data sets which can be wealthy and contain quite a lot of signals and deriving that recent information or understanding, how I like to consider it, the chemistry of that conversation is playing a really critical role I feel in AI powering customer experiences today to be sure that those experiences are trusted, they’re done right, and so they’re built on consumer data that may be trusted, not public information that does not really help drive a positive customer experience.

Getting back to your idea of customer experience is the business. Certainly one of the most important questions that almost all organizations face with technology deployment is easy methods to deliver quality customer experiences without compromising the underside line. So how can AI move the needle in this fashion in that positive territory?

Yeah, I feel if there’s one word to take into consideration with regards to AI moving the underside line, it’s scale. I feel how we predict of things is de facto all about scale, allowing humans or employees to do more, whether that is by increasing their cognitive load, saving them time, allowing things to be more efficient. Again, that is referring back to that augmented intelligence. After which after we undergo artificial intelligence pondering all about automation. So how can we provide customer experience 365 24/7? How can allowing consumers to succeed in out to a brand at any time that is convenient boost that customer experience? So doing each of those tactics in a way that moves the underside line and drives results is vital. I feel there is a third one though that may not receiving enough attention, and that is consistency. So we will allow employees to do more. We will automate their tasks to supply more capability, but we even have to supply consistent, positive experiences.

And where AI and machine learning really help here is finding areas of variability, finding not only the areas of variability but then also the foundation cause or the motive force of those variabilities to shut those gaps. And a brand I’ll give a shout out to who I feel does this incredibly well is Starbucks. I can go to a Starbucks in any location world wide and order an iced caramel macchiato, and I’ll get that very same drink experience whatever the hundreds of Starbucks locations. And I feel that consistency plays a very powerful role in the general customer experience of Starbucks’ brand. And when you consider the logistics of doing that at scale, it’s incredibly complex and difficult. If you’ve the information and you’ve the precise tools and the AI, finding those gaps and offering more consistent experiences is incredibly powerful.

So could you share some practical strategies and best practices for organizations to leverage AI to empower employees, foster positive and productive work environments, after which also all of this is able to ultimately improve customer interactions?

Yeah, I feel the general positive, going back to earlier in our conversation is there are a lot of use cases. AI has an incredible opportunity on this space. The advice I would offer is to focus first on a crystal clear, high-probability use case for your online business. Auto summary or the automated note-taking of agents after call work is becoming an increasingly popular one which we’re seeing within the space. And I feel the explanations for it are really clear. It is a win-win-win for the worker, the client, and the business. It is a win for the worker because AI goes to automate something that’s mundane for them or very procedural. In the event you consider a customer support representative, they’re taking 40, 50 possibly in upwards of 60 conversations a day during their job, taking notes of what was talked about. What are motion items? Very complicated, mundane, tiresome even. They do not like doing it.

So AI can offload that activity from them, which is a win for the worker. It is a win for the client as quite a lot of times the agents are great at note-taking, especially once they’re doing that so often, which might result in that unlucky experience where you’ve to call back as a consumer and repeat yourself since the agent you are now talking to cannot understand or doesn’t have good details about what you called or interacted with previously. So from a consumer experience, it helps them because they need to repeat themselves less often. The agent that they are currently speaking with can offer a more personalized service because they’ve higher notes or history of past interactions.

After which finally, the third win, it’s really good for the business since you’re saving money and time that the agents now not need to manually do something. We see that 30 to 60 seconds of note-taking at a business with 1,000 employees adds as much as be thousands and thousands of dollars yearly. So there is a clear-cut business case for the business to realize results, improve customer experience, and improve worker experience at the identical time. I feel as you are hopefully venturing into leveraging AI more to enhance your online business, the important thing advice I would offer is simply to deal with those crystal clear high-probability use cases and get those early wins after which reinvest back into the business.

Yeah, I feel those are the positive elements of that, but concerns about job loss resulting from automation are likely to crop up with AI deployment. So what are the opportunities that AI integration can provide for organizations and their employees so it is a win-win for everyone?

And definitely empathetic to this topic. As with all recent technologies, each time there’s excitement around them, there’s also this uncertainty of what is going to those long-term outcomes be? But I feel after we historically look back, all transformative technologies have boosted GDP and so they’ve created more jobs. And so I see no reason to consider this time around will probably be different. Now those jobs may be different and recent roles will emerge. In terms of customer experience and the worker experience one interesting theory I’m following is, if you happen to take into consideration Apple, that they had a very revolutionary model where they branded their employees geniuses. So that you’d go into an Apple store and you’ll speak to a genius, and that model carried through all of their physical flagship stores. A really positive model. Back within the day, people would actually pay money to go speak to a genius or get a priority customer support slot but a model that is really hard to scale and a model that hasn’t been successful in a virtual environment.

I feel after we see AI and quite a lot of these recent technology advancements though, that is a primary example of possibly a brand new job that does emerge where if AI is offloading quite a lot of the interactions to chatbots, what do customer support agents do? Possibly they change into geniuses where they’re playing a more proactive, high-value add back to consumers and overall improving the service and the experience there. So I do think that AI could have job shifts, but overall there will be a net positive similar to there was with all past transformative technologies.

Continuing that look ahead, how do you see the era of AI evolving by way of customer and worker experience? What excites you concerning the future on this space?

This is definitely what I’m most enthusiastic about is after we take into consideration customer experience today, it’s highly reactive. As a consumer, if I even have an issue, I search your website, I interact together with your chatbot, I find yourself talking to a live customer support representative. The patron is the driving force of every thing and the business or the brand is having to be reactive to them. Where I see the long run evolving by way of customer experiences, is being rather more proactive with the convergence of knowledge, these advancements of technology, and definitely generative AI. I do see AI becoming smarter and being more predictive and proactive to alert that there may be going to be an issue before the buyer actually is experiencing it and to take motion on that proactively before that problem manifests itself.

And just a fast example of possibly there is a media or a cable company where a tool is reaching its end-of-life state, so fairly than it have it go on the fritz the day of the Super Bowl, reach out, be proactive, contact that individual, give them specific instructions to follow. And I feel that is really where we see the advancements of not only big data, AI, but just the abundance of the power to succeed in out in preferred channels, whether that is a straightforward SMS or a high-touch service representative reaching out really where the long run of customer experience moves to a rather more proactive state from its reactive state today.

Well, thanks a lot, Andy. I appreciate your time, and thanks for joining us on the Business Lab today.

Thanks. This was a wonderful conversation, Laurel, and thanks again for having me.

That was Andy Traba, who’s the vp of product marketing at NICE, who I spoke with from Cambridge Massachusetts, the house of MIT and MIT Technology Review.

That is it for this episode of Business Lab. I’m your host, Laurel Ruma. I’m the worldwide director of Insights, the custom publishing division of MIT Technology Review. We were founded in 1899 on the Massachusetts Institute of Technology, and you’ll find us in print on the net and at events annually world wide. For more details about us and the show, please take a look at our website at technologyreview.com.

This show is out there wherever you get your podcasts. In the event you enjoyed this episode, we hope you will take a moment to rate and review us. Business Lab is a production of MIT Technology Review. This episode was produced by Giro Studios. Thanks for listening.