Experts of Experience - Agentforce: Why CEOs and Customers Are Asking For AI Like This!

Episode Date: December 18, 2024

“Limitless.” That’s how Kishan Chetan, the Executive VP and GM of Salesforce Service Cloud, describes the future of AI in customer service. Kishan Chetan explains why customer experience has evolved from deflecting customer interaction and how state-of-the-art tools like Agentforce are the key to providing proactive customer engagement, meaningful connection for employees and customers alike, and equitable accessibility for every type of customer. Whether you’re searching for that hidden, game-changing data that’s currently free-floating in an untitled spreadsheet, or you need to centralize your customer’s feedback so every department offers impeccable service, or you simply want to know how to choose, pilot, and customize the right AI tool for your business… this episode is for you.Key Moments:00:00 Introduction to Customer Efficiency00:41 Transforming Customer Service with AI02:06 The Limitless Future of AI03:48 Proactive and Reactive AI Service05:47 Introducing Agentforce07:49 AI Agents vs. Chatbots09:43 Human and AI Collaboration17:15 Real-World Examples of AI in Action22:58 Leveraging Unstructured Data for Better Operations23:20 Unified Knowledge: Powering AI with Comprehensive Data24:03 Challenges in Centralizing Data for AI25:55 Importance of Quality Data and Human Curation26:47 Practical Tips for Implementing AI in Customer Service28:16 Choosing the Right Channels for Customer Interaction29:21 Balancing AI and Human Interaction31:37 Piloting AI Solutions for Maximum Impact32:23 Creating Exceptional Customer Experiences with AI36:04 Future Trends in AI and Customer Service38:25 Potential Pitfalls and Considerations41:34 Optimizing Customer Experience: Real-World Examples44:04 Advice for Customer Experience Leaders –Are your teams facing growing demands? Join CX leaders transforming their strategies with Agentforce. Start achieving your ambitious goals. Visit salesforce.com/agentforce Mission.org is a media studio producing content alongside world-class clients. Learn more at mission.org

If we help our customers to be efficient, they will love us for it. Exactly. Don't think of just our time and our efficiency, but think of what it means for the customer. I mean, that's customer experience. Nailed it. Let's not forget, this is not AI in isolation. It's AI that works seamlessly with humans. So making sure it has the right loop to the humans, as well as the right AI that sits along with the human to make that more productive. I'm sure your CEOs are asking for it. I'm sure your CEOs are asking for it.

I'm sure your customers are asking for it. Start with something that you can turn on, show value, and then you can expand. Hello everyone and welcome to Experts of Experience. I'm your host, Lauren Wood. Today I'm thrilled to have Kishan Sheytan, the executive vice president and general manager of Salesforce Service Cloud, to discuss how Salesforce is transforming customer service with AI. I attended Dreamforce this year, and I have to say it is wild what Salesforce is bringing to the table when it comes to customer service and AI. The innovation is absolutely out of this world, and we're going to dive into all of that today. to the table when it comes to customer service and AI. The innovation is absolutely out of this world and we're going to dive into all of that today.

But before I do, I think it's important to quickly highlight exactly who it is we're speaking to. Because Kishan has spent his career innovating at some of the world's largest tech companies, previous to Salesforce, he was at Microsoft, HP and SAP, just to name a few. So we are truly speaking to an expert today

on Experts of Experience. And let's dive into it. Kishan, how are you? It's an honor to be here. This is such a fantastic set of discussions, and I'm thrilled to be here. Awesome. So I'm so excited to dive into this

because when it comes to AI, everyone's favorite topic these days, customer service is really one of the best use cases for this new technology. And Salesforce is really at the forefront of creating what I would think of as our AI future when it comes to customer service. From your viewpoint, what does customer service look like as we look forward in time and how is AI impacting that? Yeah, that's such a fantastic question. If you think about it, I think the biggest thing that I think customer service will be with AI is limitless. That's the word I like to use.

Let me explain what I mean by that. If you look at it traditionally, customer service was always focused in some sense on minimizing the number of interactions you had. It was about deflection. It's about not actually speaking to somebody in customer service because you were focused rightfully so on managing costs because every company has to manage its bottom line. And customer service, while it was very important, was something that you could manage, but instead of that, imagine that you're in a world where you could actually have a conversation with somebody at the company or speaking about

24 by seven across any channel, whether that's voice, whether that's like text or whether that's on WhatsApp that I, that I love a lot. And you could do that every time in any part of your journey. Like, you know, I'm exploring to buy this like fantastic new headset that I love a lot. And you could do that every time in any part of your journey. I'm exploring to buy this fantastic new headset that I need. And I'm trying to do that, or I'm trying to do that during the purchase or after it. And I could always get somebody on hold.

That would really change customer service. And I think that's the biggest opportunity AI kind of offers. Because a lot of people look at it as, hey, it can really help save costs, which is true. But where I look at it is it can essentially offer service 24 by 7 across all parts of the customer service journey. So it's not only reducing the cost, because we always look at customer service as a cost center. So reducing that time and effort, but also providing proactive service that actually creates a better experience overall. Exactly. So it's proactive in the sense that, listen,

if I went in and I inquired about like a headset and I moved off of it, can I actually get a reminder? Can I get some follow up from there? So it's proactive in that sense, proactive because you have a product and your telemetry indicates that. Like for example, several of the products, like a car now, has so much telemetry. It was great when I was driving my car the other day and I got a couple of notifications on my phone saying that you need to take care of

these couple of things. And that was fantastic service. I did not even know it needed to be done. It already happened in my electric car, which is connected to the internet and everything was super well connected. So that's absolutely what we see, which is proactively. But even in addition to proactive, even in reactive mode, imagine I'm in an airport, I'm in a totally different time zone, I'm calling what is late in the night in the US time, and I actually still speak to an agent on the other side, a voice agent, which is really really an AI powered agent. And I can have that conversation anytime versus being told you have to call between 9 p.m., 9 a.m. and 5 p.m. on regular working hours in US time. So I think that's the difference. It's proactive and also like just the ability to get it 24-7. Yeah. And just easier for the customer. I think this was the thing that stood out to me so much

when I was at Dreamforce and seeing the Agent Force demos was that you can pick up the phone and you can call a company and you get someone, an AI agent answering the phone immediately. You don't have to say press one, press five, press, nah, nah, nah, nah, nah, to get all the way through. You're just speaking to someone normally in natural language and getting your problem solved quickly and efficiently.

And I was honestly, I thought that this type of voice AI customer service was something that was like, I don't know, three, five years down the road, but. It's here. It's here right now. And it's really amazing. It's really amazing. So I want to talk about Agent Force today quite a bit, but could you just tell everyone for those who maybe aren't familiar with what Agent Force is, can you give everyone a bit

of a download on what Salesforce has brought to the world? Oh, that's fantastic. Yeah. I mean, Agent Force is really a key part of our overall kind of platform. It's built on our overall platform. What Agent Force does it, it makes it very easy to build an AI agent across any role. So you could have an agent for service, which is what we're discussing, but it could be for sales to help with better sales across any industry.

It could be for retail, could be for consumer goods, financial services, and across any product. So that's kind of the core force, part of agent force. And what you might ask is like an agent. And the way we look at agent is, agent is somebody who like, which has a role, like you're selling or you're marketing or you're servicing, it works across a set of channels. So voice, as you mentioned, could be on a text, could be on a chat. It works on a set of channels. So voice, as you mentioned, could be on a text, could be on a chat. It works on a set of data, which is your data. As a customer, it works on your data,

so it notes what's there in your knowledge base, the agent knows what's in your website, et cetera. And it essentially drives a set of actions, and that's crucial, because really what the agent is focused on is taking this data and converting that into action. The key part of agent forces does that in a very intelligent way because at the heart of all of this is what we call our Atlas reasoning engine. It basically takes all this data,

processes that data, reasons on it, and creates this output and orchestrated set of actions. What that translates into is I call up somebody saying, listen, I'm coming in a little late today into my hotel, which I do all the time because flights seem to run late. Can it change my check-in or have my mobile key ready? I call up a bank and say, listen, I want to change my address. Can it actually make that happen versus putting you in a queue? So that ability to take data and drive actions

is what agent force really drives in an autonomous way. How is it different from a chatbot? Yeah, the biggest difference from a chatbot is chatbots are typically pretty rigid, right? So it's rule-based. So you basically built a lot of rules and you've said, hey, listen, if this happens, then do this. And if this happens, do that.

And that's all great, right? And it's a great way of like driving automation in the first place versus anything else. But it's still rigid and you can run into walls very quickly. So quite often when you work with chatbots, you run into this thing that I don't know how to react to this. Now you need to speak to a human representative, right? That's the very common thing. But with the AI agent, it's powered by LLMs. So that means that the whole interaction is far more conversational. Its reasoning goes well beyond a rule-based system.

It truly has the ability to reason based on all of your data that's in your system, whether it's your order information, whether it's your knowledge or this documentation. And then it drives these actions, which can be very broad. And so that's the combination of all of them. And then we built this thing in Agent Force

where you can kind of train this agent to be focused on specific skills, address a set of topics and drive actions. So it's a lot more open and a lot more intelligent than the chat bot, which is more rule-based. I think everyone listening has probably had an experience with a chatbot where you're saying you're asking a question and it's misinterpreting you. And then you ask it again and it misinterprets you again.

And then you're just like, human, human, send me to a human. That's exactly right. And it is so painful. And so that is why I really think that, you know, it is amazing for anyone in the customer experience space to know that it is possible for us to transcend that customer pain using AI to help us create experiences that feel human to human, but also aren't.

And I want to talk about that human to human connection because there's a lot of pros and cons here. In some cases, there are not enough people on the team to answer the phone immediately. An AI agent will always answer the phone immediately. But then sometimes the AI agent can't handle everything that a human can and we need to do the handoff. And so tell me a little bit about the role

that humans play when it comes to an AI agent and how we really create that relationship that provides an exceptional service. Yeah, and that is super crucial and something that we think a lot about, Lauren, that's absolutely crucial because listen, there are reasons why an AI agent might not be able to handle it. That's because there are things that just need to go in once in judgment. What if a customer sounds really pissed and angry? You perhaps don't want to have that in that conversation or because of your process, if that discussion is about buying a new product or doing more upselling, you might really need to have that human connection in there.

So your business need might drive that human connection. So, but what becomes really important is to pass the full context and have all of that context always available for the human. So, which means like a very simple thing, how often have you had an experience where you've entered all this information in a form or with a

bot, and then you go into a person, they're asking you the same questions and you're like, why didn't you know this? I've already entered it like two times in this conversation. Are you seriously asking me for my email? Are you crazy? So I think that stuff's like the thing that we should not do, right? So you should make this experience like super seamless. So that way, when you go in, you know, all of this information, you've taken on everything that you've had from the bot or the agent in this case. And that's available to the human agent to do really well. And then we also have agents that sit alongside the human. Like, for example, how can like, how can you get a prescription on like,

what are the three things you need to do if somebody is complaining about a bad router at their house? That agent essentially makes the human as well far more powerful. But the most important thing is any conversation that's happened before with an agent, pass the full context, provide the right context to the customer, and most importantly, the customer should just feel it's seamless. You know what?

From a human, you might actually go back to an agent for some other reason, or you might go to a different channel because you might have started the whole thing in a chat message, but now you're going into your phone and we really don't want you to chat on your phone when you're driving. So you might just convert that into voice. So all of that needs to be very seamless. And we've put a lot of effort to make sure that it is. I think you mentioned a really great point

that the AI agent is not only for the customer, but also for the employee. And we talk a lot on this show about the importance of the employee experience. And if we can make it easier for if the customer does need to speak to a human, how can we enable that human within our company to provide a great service easily, efficiently,

so that they can spend more attention on the customer that they're speaking to rather than going into multiple tools or systems to find the right information and propose the right thing? Like, it really is, the AI agent is really acting as an advocate or as an assistant for the human employees to make sure that they're able to do their job better, faster, easier, all that great stuff. And that's exactly right, Lauren. I mean, listen, I mean, if you look at the average

called contact center, and you know, I was speaking to two customers this week. One of them, they basically hire 120% of their contact center every year. So that means that they lose all of their apps and they have to hire 20% more. So like how much can you actually learn about your own product with that type of a churn? And then I spoke to another customer who was like, you know, had very low, you know, attrition, but they were moving from one department to the other. So they would be doing travel one day, then they would be doing luxury the other day in

this large retailer. So then it becomes very important for these like agents to be very knowledgeable. And what you don't want on a call is somebody telling you, and this has happened to all of us, I'm sure, wait for a moment, I need to go like, you know, go check for this and I need to go do research about this or I need to go speak to my supervisor and you're like, you know, goes check for this and I need to go do research about this or I need to go speak to my supervisor and you're like, why am I calling this place if these people whom I'm talking to are not the experts? So I think in some sense, the agent sitting next to you makes you the expert because everybody's short on time. We all want our answers now and when we want it and having this agent providing you that and helping that human employee provide that right answers

with the right context and being able to do the automations makes it super important because then as you said, your focus as an agent is on building the relationships, helping drive growth, helping advocate the products, helping advocate the services, which is way more important than trying to like find some information so that I can actually like solve the problem for my customer immediately. Completely. I had an experience the other day where I called my healthcare insurance company, and I've called them many times in the past. And it's usually this thing I put off forever because it's painful. Like it's waiting on hold to wait on hold to wait on hold to get handed to someone else to wait on hold again.

And it's just like I just it's hours of my day and I hate it. And I called them and it was an AI agent that answered the phone, asked me a bunch of questions. I had a conversation with this AI agent. She handed me to a human and immediately a human picks up answers my questions, says like he knew everything. I actually didn't need to see if there was Salesforce customer because it was like, it was amazing. I was like, I actually want to call my insurance company because I, this was so easy. It was so seamless. And I just, I value that company more now rather than them being a pain for me. It was actually a, wow, they are

acting in my best interest. They are considering my time and they're giving me the attention that I need, doing the exact thing for me that I'm asking them for. And I can walk away with a smile on my face. And I think that that's the big difference when we talk about like a great customer experience versus a lacking customer experience is how quickly are we able to give the customer an answer? How easily are we able to guide them to get the help or the support that they need? And I think it's just one of the biggest pain points in the customer experience space is really the lacking technology that we've had. And like

you said, losing all of your employees every year, because probably there's pain for them in doing their job. We can solve that with this new technology. And it's just, the future is so bright. I'm so excited. Yeah, no, I mean, it's great. I'm glad to hear that you had a call with your health insurance company and you left with a smile. I mean, that's exactly what the insurance companies want as well. So that's fantastic. Totally. It's like such a rarity.

And I was like, oh, this is so good. I'm so happy for this. Say goodbye to chat bots and say hello to the first AI agent. Agent for service agent makes self-service an actual joy for your customers with its conversational language anytime on any channel. To learn more, visit salesforce.com slash agent force. So I want to talk a little bit about some examples of customers that you have or just

like ways that this technology has been used and what kinds of tasks? We talked about a couple of them, but I'm just curious to know a bit more of the breadth of the different types of things that these AI agents can do to help us. The beauty of Agent Force is given it's built on the platform, any type of automation that you've done with Salesforce,

you can essentially do that with Agent Force. So think of that. And let's take a few examples. And I like oversimplified it to essentially bring it to kind of three categories. Like the first one is what I call action-oriented. So think of all the customer service interactions you've had to find like a quick piece of information.

You called up like a FedEx or you called up a UPS and any of these, and you want to know, hey, where's my thing that's going to get delivered to me? I want this quick piece of information. It's very transactional. I want to go get that. So the ability to have that be a voice call or any like, you know, agent interaction and getting that information and doing that quickly, that's like what I call extremely like action

oriented. The second one is what I call more order-related. It starts with action-oriented, I want to know about my order, but very quickly it starts to move into, hey, can I know more about this product? Can I actually change my order? It starts to become more consultative. It starts to become more understanding what this product's about. That is where you could start with an agent, you could get that scale,

but very soon you realize, hey, look, somebody is here trying to buy this luxury bag. And at that point of time, we had this fantastic example where Gucci did a lot of work with us, where you quickly want to go to a human, and the human needs to have the right message about the product, the right message about the brand. And that becomes very important. So that's where the agent and AI have the right message about the product, the right message about the brand. And that becomes very important.

So that's where like the agent and AI in the flow of work really helps that, you know, helps the human agent be far more productive. And the third one is what I call like very knowledge-centric. So if you're like reaching out to a high-tech company and you wanna understand like, how do you fix this particular problem

that I've had with my router? Like, my router? Cisco is a great customer of ours. That requires looking at all of this knowledge, looking at all of this data that's there in Salesforce, outside of Salesforce. That requires a lot of work on unstructured data. That's another great example where agents look at all this unstructured data, synthesize it back into two or three things. We've had customers who were spending almost $100 to $200 on a customer conversation because it just took that time to go get that information. Imagine if you could save that time and make that much cheaper because you don't have to read 300 or 400 documents. Those are examples. Very quick, get me an answer, logistics, hospitality, those

types of industries. Great examples, like we went to OpenTable, you can go reschedule appointments, find information about restaurants. That's super effective with that. Very action-oriented. To the second, which is more order-oriented, like I want to find out more about this product. Do you have this great cashmere sweater for us, like sacks on a retailer? How can I return it to the final one, which is like, hey, listen, I'm having this problem with the product that I have at home. How can I go change it? How can I

go set up my best music system? So those are all examples of customers using Agent Force, doing all of these different types of actions. I love it. I want to talk about data. You mentioned data, which is a big piece of this. And being able to really process lots of unstructured data. I mean, any customer experience leader I have ever spoken to has always said, we are sitting on a gold mine of information about our customers. But the challenge is then actually processing it into insights that we can utilize to be more proactive, to understand our customers better.

How does Service Cloud in general help organizations to really process this data to make use of it? Yeah, I mean, that's such a fantastic question. I mean, there's like so much different types of unstructured data, right? So if you look at like unstructured data that sits in, let's say, a conversation you've had with a customer, like a phone call, a live transcript of a phone call or a chat, there's a wealth of data. A lot of our customers have gone back and tried to do maybe a survey with a customer to understand how they feel and think. But you have this wealth of information that you can use to actually understand what

the customer is thinking without asking them. So, imagine if you could use all of that information to understand the sentiment, understand the intent. Are they looking to buy it? Are they unhappy? Are they looking to return? And you can use that to do aggregate it, like give an offer to somebody who's called in an airline because their flight was delayed or their baggage was damaged and give them an offer, give them the ability to get a better discount on their

next flight. That would be fantastic. And that's why we have this customer experience intelligence to be able to look at this aggregated intent and sentiment. Imagine you're having a whole bunch of calls around something broken in a fridge. If you're a manufacturing company, there's a certain problem with their clients, and you're seeing 30 or 40 calls coming to your contact center. What if you could take those different cases and make it a knowledge article,

so people can easily find this information online. But more importantly, as you said, proactively send out a notification to everybody who has that machine, a washing machine or a dishwasher, and ask them to go fix it. Fisher-Pakel is a great example of a customer of ours who was in the Service Cloud Keynote. They can use a lot of this asset-based data. That's another great example of where we're using all of

this information coming into the contact center and being able to automate it, better operations, drive more knowledge, or deploy more agents to focus on a specific area because that product line is getting a lot of questions and a lot of issues. Mm-hmm. So that's another great way of using data. And finally, another place where unstructured data becomes super crucial is all of this knowledge information.

Imagine if you could get knowledge and knowledge information from your SharePoints or from your OneDrive and from your files or from your websites and bring this all together and use that to ground your AI. That's where our unified knowledge and all of the investments that we've done to bring knowledge from everywhere else really powers it. So fantastic investments in summary on analytics to understand sentiment intent, to manage your operations of your contact center using unstructured data, as well as using unstructured data like knowledge to drive automation in a much bigger way. So those are the big places where we use unstructured data in a much nicer way. I think that this is one of the places that most companies struggle with a lot

is actually bringing all of the information into one place so that AI can make the right decisions or guide you in the right way considering all factors. Because it's not only what are the customers saying to you, right, it's what are they saying to you via your customer service channel? Might also be what they're saying out there in the world on social media or in focus groups or in industry research,

all of these things, we need to be taking it into consideration. And I think that that's one of the areas where, as I talk to people just in my own work about implementing AI, is we really need to look at the number one thing is what information are you training that AI on and does it have all the information?

And it's one of the biggest challenges is getting all of that information into one place where it can actually be processed. And I'm so glad that Salesforce obviously is thinking about that because that's what you guys do. But it's really a holistic approach to being able to provide the best service. Yeah. No, Lauren, thank you for that. Yeah, I mean, absolutely. That's a huge part of it,

which is how do I get all of this information, this wealth of information that's there? I mean, imagine if you're a high-tech chip manufacturer, I met them a couple of months ago, super complicated product. The people who use it are these highly technical engineers. And it takes them 10, 12 hours to go through all of these documentations and understand how to do something.

Imagine if you could get all of that document together so that you could answer a bunch of questions in a much easier way. And that's where bringing the knowledge together helps. But there's a second part of it as well, which is I was speaking to another high-tech company, which is like focused on like better like networking app, like making your networking much nicer. For them, what was important is not just having a lot of data, but having the right data. So making sure that you have the right quality, that you didn't bring a

bunch of like, you know, garbage data, that you didn't bring a bunch of garbage data, because in some sense, it's garbage and garbage out. So you brought the right data. You already had the right quality. So that's where another place where data is super crucial, but having that human element to make sure that this knowledge is rightfully curated, is updated, it has the right segmentation, the right categorization of it

becomes super important. And that's where it's really around bringing all of the right segmentation, the right categorization of it becomes super important. And that's when it's really around bringing all of the data and making sure that you use the right part of that data to essentially power your AI because otherwise it just won't be able to do much. Garbage in, garbage out. I hear this all the time as I'm talking to folks about how do you get the right data to really power your AI? What advice do you have for leaders as they look to implement AI in their customer service

when it comes to really getting the right data? Any tips or tricks for our listeners? Yeah, no. I mean, listen, when I have AI discussions, and I've had AI discussions over the last several years across different industries, it always first comes back to a data discussion, which is, do you have the right data? Do you have the right things in place in order to do that? So I think that definitely becomes a huge part of that discussion. So that's one thing that I always kind of start people with, which is, like, you know, do you have the right data? What is the right set of data

that you want to take? But while I say that, I also am mindful of the fact that you don't want to make this a massive data project, because there's a lot of value that you can get right now with AI. So it's essentially about figuring out what are the use cases that you can immediately turn on that you can get value. Can I make it very easy to find answers about a set of knowledge articles? Can I turn on an agent to answer a set of questions? Can I turn on an agent for scheduling something? So data is super crucial.

But if you identify the right use cases, you can already figure out the right data that you need to get for that use cases. So be pragmatic, figure out where you have the highest value, the highest pain point that you can immediately address customers' needs. And more than that, delight the customers and get the right data for that. The other part of agent strategy and AI strategy very quickly becomes about channels. How are customers trying to reach you? If I'm trying to go turn on a project in Latin America, maybe I have some fantastic retailers, really great telco companies who are our customers.

And for them, WhatsApp is absolutely crucial. Business is conducted on WhatsApp in Brazil and Latin America or in India. So having that channel on is super crucial. If you're a much more complicated high-tech company, like I was crucial. If you're a much more complicated, high tech company, like I was speaking to, not just a high tech, I'm speaking to like a healthcare company, which has a fairly complex product, and they do patient services with their patient. Most of their questions are on email because they're complex, they're lengthy, they're like, you know, you need to have a bunch of questions. So having the right channel mix turned on for what you need, super transactional, maybe an SMS, maybe WhatsApp, questions. So having the right channel mix turned on for what you need.

Super transactional, maybe an SMS, maybe WhatsApp. For the geography, the right mobile channel. If it's more relationship-oriented, the demographic that you're targeting is slightly different, voice becomes super important. So having that combination of the right channels, having that data, having that use cases. And finally, let's not forget, this is not AI in isolation.

It's AI that works seamlessly with humans. So making sure it has the right loop to the humans. If you want to create the loop between agents and humans, as well as the right AI that sits along with the human to make them more productive. So then that's, that's kind of the way to think about it. But, you know, my advice to leaders would be there's so much demand. I'm sure your CEOs are asking for it. I'm sure your customers are asking for it. Start with something that you can turn on, show value, and then you can expand. Yeah. Taking it one step at a time, I think is... I hear a lot of companies say like,

okay, we're going to overhaul everything. Like all of our tickets are going to end into AI tomorrow. And I think that's a dangerous approach. Oh, yeah. Exactly. Because we do need to find that right human to AI balance. And I'm curious to know if you have any tips there around how to really navigate that in what areas is the human involved versus what areas does the AI kind of run on its own? How do you approach that with your clients? I think the way we've looked at it with clients is one way to think about it is what are the areas that there's actually a lot of demand, customers are not being satisfied, they're not getting the answers the way they want to, it's essentially not being serviced. So

that's literally not moving it away from humans to AI, you're essentially going after almost a greenfield area where you could go solve. So that's one way of thinking about it. The second way is when you look at your kind of user base itself and you know, you're tier one, tier two in your contact center, how much of that is like repetitive things, like, you know, is it always about resetting your passwords or updating your accounts? So what are the repetitive tasks that you can make sure

that you have a very well-defined process that you can automate away and the customer will actually be more delighted because it happens faster, because they don't have to wait for somebody to become free up. So I think having that ability to like, make sure that you have the right process

with the AI supporting the human to then being totally autonomous becomes very important. So starting them in the green field or start with something that you just have a lot of repetitive cases around. And then the third one is I always like to pilot it. Pilot it in specific regions. Pilot it with specific product lines. piloted for specific processes, and see value, show the value to your people, your leaders, and start to expand. So that essentially is the best way to do it. So my motto is, that's what I tell customers,

turn on something and show value. Make sure that you're comfortable, because it's in front of your customers. It doesn't have the right, too many hallucinations. And our technology is built to like minimize hallucinations. Make sure that's accurate and solves the customer's problems and then expand. Yeah. The green field area.

Let's circle back to that for a second, because there is, once we find these efficiencies with AI, once we're able to reduce the workload of each employee to actually service the customer, and we move away from the, let's deflect as many tickets as possible, we can actually handle them. Now there's this space for us to create great experiences. And it's just the area that I'm the most excited about, because now we can actually look towards,

what do we do with this extra time? How do we go above and beyond? How do we, like you had said earlier, proactively look at if something's not working, or if someone complains, or they don't have a good experience, we can go and proactively say, here's an offer, or here's something towards your next experience. And I think that there's just so many opportunities.

I'm curious to know what you've seen out there in the market in terms of what companies are doing to really make the most out of this extra time that they have and provide a great experience in turn. Yeah, I mean, that's a fantastic question, right? I mean, one of the things that we saw is a few things that people are doing with extra time. So we have this great example with Formula One. All of us know Formula One. They had a race recently, a couple of races in the US.

Initially, their contact center reps used to handle a lot of like repetitive elementary tasks. As that got more and more automated away with AI, they started to handle things which were much harder, much more difficult. So one thing is that you're just handling far more, far many more customer reaches that you never had before. So that's literally like instead of handling thousand cases a week, you're just handling like, you know, five, ten, depending on what it is. So that's a great example of where you're using the same number of people, but you're essentially like handing a lot more than you ever had before. Yeah. If you look at companies that, you know, there are high tech companies, they're growing 10 to 15% every year. They have, you know, that many more customers for them using this AI was a way to

handle the load with the same number of people. So those are examples of like, you're essentially being able to do just more. The third category, which is like the green field, is where we are seeing a lot more kind of focus on, let's have the service reps start to kind of build relationships, drive a lot more growth, be an advisor. So if you're in the retail industry, for example, especially in luxury retail or you're in wealth management,

that's what you want your agents to do. So maybe you're seeing that moving away from transactional things like update my account or where's my order. If I can get like a wise agent to handle that, then I'm focused on advising people about the latest, like look, what is the latest bag? What are the materials in this? Are these sustainable resources?

The ability to advise people if you're a luxury company with the right bags or what we've seen, that's where I think we are seeing using that time to essentially take on more of the revenue growth roles, if I may. That's definitely something that we are seeing in a much more structured way. The building of a relationship of really being able to say, I actually have a little time, I can ask this customer questions. I can better understand what it is they need so that I can show them value based on what it is that they're looking for. It opens up a whole new opportunity for what customer experience means, which is really exciting. I want to spend a little bit of time here talking about the future and just your opinions. You've been in this space for a long time.

You've helped to build some of the greatest tech companies in the world. And I'm curious to know what you think as we look towards 2025, what emerging trends or technological shifts can we expect? Yeah. I mean, I think like a few things that I see. I mean, I think this 2025 is going to be where you see a lot more agents used in a lot of scenarios which go well beyond service. I mean, even at service, we are scratching the surface, right? Starting to have agents which do more. I think that that's definitely going to be one big thing. The second is, I think the area that,

you know, we both spoke about this is around proactive. I mean, like most of the companies, like 80% plus of companies believe they're proactive, but when you turn back and ask the question to, you know, customers, they only think 30% of the companies they work with are proactive. So like bridging that gap and, you know,

being far more proactive, far more delightful in your experiences. I want everybody to have the same experience as what I have with my electric car, where it tells me that I need to go do this. That's exactly what you need, right? So I think much more proactive, but based on data, I think it's going to be a big part of the trend

that I see for next year. And then in terms of customer experience particularly, I see that we'll see a lot more customer experience and service in the entire journey. You'll have a lot more pre-purchase because you really can do it now at scale. You'll have a lot during purchase and post-purchase. So I think we'll see a lot of that across the board. Now, I also think that we will start to see some questions from different countries and different geographies about what's the impact of this.

And we'll have a healthy debate and discussion around that, I think, as we go on. But this is one where the customers, me as an end customer and end consumer, will just be delighted by the type of service that we'll get. I really want everybody to have the same insurance experience that you had when you were delighted at the end of it. That's what we need and we'll see more of that across the board.

I'm an optimist. Great. On the flip side, I know you're an optimist, I am too, but is there anything that we need to watch out for? Yeah, I think what we need to watch out for is, this is a machine that can go really fast and can really support a lot of cases. Are we making sure that we have the right monitoring of that?

Do we have the right oversight of it? Are we, as a company, as a brand, if I can go create this experience for everybody, is it diluting in some sense like what I stand for? Who are my core customers? What's my vision? So making sure that first it's right oversight, it's solving the right problem. It's not like having hurting the brand. I mean, like we've heard these examples of this airline company, which had a problem in terms of a suggestion and I don't want...

Brands have to be very careful about that. And second, making sure that the brand continues with what its core purpose, what its core value is, what its identity is. I think that's important. And third, we need to make sure as a society that as we are using this to handle a lot of what would be done by labor, how do we ensure that the employees are then trained to take on this additional building relationships,

driving growth. So there's a whole bunch of like training that we kind of broadly need to do. So there are things that we need to concertedly watch out as a group. I think that employee training piece is something, I know a lot of people are worried. They're worried about their jobs in the face of AI because AI can do things that they're doing currently.

And it kind of goes back to the values piece as well that you mentioned for companies to think about what is the experience that we want to provide to our customer? What I find often is that what companies want to be providing is not what they're providing. And I think if we stay focused on creating the experience, that we really envision for our customers, and we allow our employees to grow into providing that with the help of AI,

everyone is going to be happier. So this is, again, my optimist view. But I really see there's such, there's an open green pasture for great customer experiences that so many companies have not been able to step into. And we can now do that. And so it's not about having less people working for us per se, it's about having more people doing the right things that create the right experience that really increases our relationship and the value that we can provide our customers, which just makes everything better, in my opinion. Couldn't have said it better, Lauren. I think that's the key thing. I mean,

they just like think of a small mom and shops store where you kind of you have this corner store where you want to go buy some flowers or things like that. I mean, now they can essentially have like an engagement of far better kind of service than they ever did before, before they were just constrained based on resources. So in that case, it's not about people. In fact, we are actually driving more business with them. So I think it opens up a whole bunch of opportunities that we just never had before. And the experience and us as consumers are just going to be richer for it.

But there are things that we of course have to work through. A hundred percent, a hundred percent. Well, I have two last questions for you. These are questions that we ask all of our guests. The first is, I'd love to hear about a recent experience that you had with the brand that left you impressed. What was that experience? I've been traveling a lot in the last few weeks and I've really had to use OpenTable

quite a bit to discover, you know, like restaurants and the type of food, et cetera. I really enjoyed kind of the agent experience that we have with them and, you know, glad to hear that we power them. And that was something fantastic because what was important for me was like, you know, I was having a conversation in a different city, then I moved to another city, and it kind of remembered that conversation and it remembered the context and remembered the taste and kind of was making better recommendations.

So I think that was like a really delightful experience. And another experience that I had was not like a brand, but it was actually a state department licensing here, where I wanted to go schedule an appointment. Who enjoys going to a DMV? I don't, right? But man, it was so efficient. Like I could easily schedule my appointment.

I got there in 15 minutes. I got done and it was all great. And the whole experience of like self-service scheduling without actually having to do anything and being super efficient on time, getting it done and getting out. That was remarkable. So I think a very nice self-service experience in one case from a government, from a public sector, there are a lot of public sector agencies using Salesforce and OpenTable, which was there in our service cloud keynote with the experience of like engaging and getting that fantastic like, you know, reservation

experience and discovery experience was awesome. On the DMV thing, them caring about your time goes so far. Feeling like a company cares about the amount of time that you have to spend. And I think the reason why people get frustrated with the DMV or the IRS or whatever is because of the time that it takes. And we're like, why do I have to give so much time in order to do this thing that you're making me do? So if it becomes easier, and I think every company can learn from this, that if we help our customers to be efficient, we are also efficient at the same time. But if we help our customers be more efficient,

they will love us for it. And it's one of those things that's just, it's, it's easier said than done. But I think it's just a really great area to focus on is how can we think about the effort our customers have to put in to do the thing that we need them to do. Exactly. It's don't think of just our time and our efficiency, but think of what it means for the customer. I mean, that's customer experience. Nailed it.

Yeah. Last question for you. What is one piece of advice that you think every customer experience leader should hear? Think of exactly the right customer you want to design your customer experience for. Who's the customer? What are you designing your experience for?

And then make sure that you have the right data for it, you have the right channels that they're on, you have the right ways to reach them. You can have the best experience design, but if your customer is looking to engage within a different place, it doesn't work. So identify your right customer, understand what they want and be where they are with the right data. Be where they are.

This is such a key thing and different customers have different preferences too. Exactly. Different customers, different preferences too. Exactly. And we need to give them options. Different customers, different segments. That's exactly right. Well, Kishan, it's been amazing having you on the show. Thank you for coming on and sharing all of your knowledge with us. The future is very clearly bright for the world of customer experience.

And thanks so much for helping us to make it happen. Thank you, Lauren, for having me and for the wonderful conversation. It's fantastic to hear your insights and I love your simple distillations of the key truths here in customer experience. Awesome. Well, I hope you have a wonderful day and we'll talk to you soon.