We Know That We Need to Know What We Don't Know

Transcript

[00:01] Steve MacDonald: Welcome everyone to the Contact Center Perspectives podcast. I’m Steve MacDonald, your host. Today, we’ve got a super interesting guest and conversation; we have on here, Ilai Szpiezak. Now, Ilai, you’re not only the CEO and founder of Dolphin AI, and I want you to explain a little bit more about that very quickly here.

[00:17] Steve MacDonald: But we’re going to be talking today a lot about data, and we’re going to be talking about that we know that we need to know what we don’t know. And in bringing that to life and how those insights literally can transform a business, and where a tremendous amount of that data and that insights comes from the thousands of interactions that we have in our contact center. But the highest voice of authority in any business or customer, right? So we’re going to get into all of that. But before we do, I’d love just to hear a little bit more about you and your background. ‘Cause I know you didn’t start out in tech; you’re not like a big tech guy, right? And you’ve got some really interesting parts of your personal life.

[01:00] Steve MacDonald: I’d love to have you share with us here.

[01:02] Ilai Szpiezak: For sure. I’m so excited to be here, Steve. Thank you so much, and I’m very pleased to share a little bit more about Dolphin, about my background, and about this area of data, which I think is one of the most important ones and powerful ones for any company. So I was born in Argentina—very quick background—half Israeli, half Argentinian. I moved to London, UK, in 2011. And I have a background actually not in tech, exactly like you said. And, I started out as an event and theater producer. I had an event agency where I’ve done over a thousand events all around the UK and the world—anything from private events to tech conferences, to hackathons, to weddings. And I learned a lot about the similarities once I moved into tech, about how similar it is to build great experiences for customers, that in one time it was about an experience where you were physically in an experience, but now it goes into the digital world, and we are building products that should delight people. I’m a big believer about customer love. I was doing events and the pandemic appeared, and I completely lost all my work, and I did one of those pivots where I got into tech, and that’s the time that I founded my first startup, which was a dance platform for dance classes online. Dance has always been my passion. I’ve danced all around the world, teaching workshops, choreographing for people from all different kinds of countries. And that’s how I got into tech. Eventually, I ended up founding a startup called Dolphin AI, which you very well said before. And now what we’re doing is building a voice of the customer AI tool that helps aggregate customer feedback that comes from all these touch points: customer support, customer success, surveys, reviews, tickets, chats, and being able to decipher what are the biggest pain points, validated by the voice of the customer, to close that loop with the product and other departments. So that’s what we’re doing.

[02:59] Steve MacDonald: I love your background and experiences—physical and digital experiences—your creativity as a ballet dancer and instructor worldwide. So I’m going to ask you, take all of that background and from your point of view, we know that we need to know what we don’t know. What is the biggest part of what we don’t necessarily know? Like, we’re not mining, we’re not bringing out that voice of the customer. Start us on this journey. What’s the most important part?

[03:30] Ilai Szpiezak: So for me, it always comes down back to thinking about the customer perspective, and there is a reason a customer reached out or reaches out to the company. There is a physical reason that practically someone wants to talk to the company, fix something, and unblock something. In the end of the day, it’s about solving for a problem I’m trying to solve or that I can’t solve. And this is how we need to understand, from the company perspective, what is this data telling us? And I think one of the biggest fails from companies is not realizing that the data is already there. We don’t just need to go and ask for customers to give us their feedback and think about, give me data, I have this data. It’s that we are constantly thinking about how we service them, not how we use the other side of the coin, which is the data that they give us when they are trying to get some help.

[04:28] Steve MacDonald: That data, we know it’s there. Those conversations are happening, right? But we haven’t always had the tools really to mine that data and understand, ’cause not all data is important data either. How do you think about the challenge of filtering and literally routing these insights, this knowledge? And how do we do that?

[04:48] Ilai Szpiezak: So, what is really interesting, and one of the insights that I learned as a founder—and I think that my job as a founder is to constantly learn and try to implement that, and that’s the job, this is the primary job of a founder—is that not everything is an insight for everyone, and different people care about different things. So how do we know what each persona, each team, each department in an organization need from the data? Because what is an insight for X is not going to be an insight for Y. And if we are able to, first of all, organize this data—because it needs to be organized, saved somewhere, joined together—it’s very different to look at a survey response, to a ticket, to a call, to a review from someone. All these areas are connected to that customer experience and the service that we need to provide to the customer. So if we can mine all data, first of all, put it in one place—that’s step one, to bring it somewhere. Step two is to be able to organize it, to analyze it, to segment it, and segment it based on data, based on customer, based on what each persona wants to listen about. And then we need to be almost like a postman, I like to say, because you are taking the letter and putting it in the right mailbox based on what that insight needs to be given to someone. In the end of the day, it’s about actioning on this data and discarding the noise and focusing on what matters now. And that changes from a quarter to quarter, month to month, persona to persona. So we need to be the right postman, or use the tools that help us act as a postman, and influence the right decision-makers based on this data.

[06:23] Steve MacDonald: So how do you do that? The analogy of a post office, right? Because, right, it’s like, “Oh, this is a letter that needs to go over here. This one goes over here.” Very visual. I love that. How do you do that? I get that when there’s zip codes and things on addresses tell us what to do, right? How do we know what information is important? How do we know what information needs to go where? Because that is a big part of, right, is getting the right info to the right people. So how do you think about that?

[07:09] Ilai Szpiezak: So the way I think about this every day, while building a tool that helps companies route the right data to the right person—and actually, we need to quantify this data—I think it’s like we need to be creative in how we think about this, because it needs to be always connected to what the customer said, because that is actually the reason, but also needs to be connected to who will care about it. So, for example, a marketing team would be very different in what they want to listen from these insights to what a specific product, a growth team, is working on: the activation of the customers or the onboarding flow of the tool, or maybe what a finance team is working on in terms of billing and payments. And all these issues appear from those contact centers and those support chats and tickets that customers are reaching out for when they want to upgrade, when they want to churn, when they have issues with changing the email, and all these are the elements that we see while building Dolphin, how we can decipher what the customer said but then allocate the right label. I call it “signal,” by the way, because it’s a signal, it’s not definite. It’s a signal that we need to listen and then make a decision. But how do we allocate the right signal based on what the company’s currently working on for each team, essentially? So it’s about having this thought process of where it comes from and where it should go.

[08:33] Steve MacDonald: So here’s a huge takeaway from what you just said, right? So I have a big background as a marketer. I’ve been a chief marketing officer. I’ve been a founder of my own company as well. One of the things you have to do is you have to know your customers, so there’s a lot of contact center leaders—there’s chief customer officers, chief experience officers—that are on and listening to this podcast. If you don’t know your internal customers and what their needs are, their challenges are, what they need to know, then you won’t even know what signals to look for, right?

[09:05] Ilai Szpiezak: Exactly. The first thing I say is when someone asks me, “How do we start this even without the tool?” is: don’t do anything but go talk to your internal people to understand what they would like to understand from their customers, because as the company grows, there is more distance between those teams and the customer, which is a gap that we need to fulfill. And that’s where all these tools, like Dolphin, I believe, can help shorten that distance. And there’s a great story about the founder of Wix, the Wix website, that he says the job of the company is to shorten the distance between what the team does and what the customer thinks and wants, because then it generates more people who buy into why they are doing something for the customer: what is the impact of that job in someone’s life? So, yeah, I always think about that.

[09:58] Steve MacDonald: That’s great, because it’s something that, we just need to get a mindset, no matter what part of the business we’re in, the customer has to be our focus, right? And then, our second customer—that’s the external customer—as contact center executives, we have internal customers, they’re every department. And the very nature of going out and asking those questions is super important, not only to understand the signals that we need to divert over to them, but just to create that connection. Right now, there is a huge issue in the industry that contact centers are considered cost centers. They’re not value centers. They’re like, “They’re necessary, we have customers, they have questions, of course we’re gonna need to answer those,” right? But it’s a cost to contain. So leaders look at cost centers, not value centers. So as a leader inside of the contact center, if we can go out and start saying, “Hey, what would you like to learn? What would you like to know?” we’re basically saying, “Look at all the value we can provide to you.”

[11:04] Ilai Szpiezak: Completely. And that brings leverage, but brings also value, because now we’re talking the language of someone else. We are transforming a contact center or support center that is seen as a cost center, like you say, into a data center. Data equals value, equals gold, equals decision making, equals valuable insights that can make us make the right decision. And I’m passionate about this while building a company in this space, that we need to transform these customer-facing teams into data centers. And we are now in the era where we are all trying to think about how AI is cutting costs, right? We can now think on the other side of how AI can bring more value from the data perspective.

[11:49] Steve MacDonald: So let me ask you about a slice of that value, because it’s very easy to say, “Oh, customers have questions on renewals, or they have questions, they can’t get this to work right or that to work right.” But a lot of time, customers are contacting us because they’re actually evaluating what we’re selling. Like, the contact center is a pre-sales organization. The contact center also, as they’re using a product, we have the opportunity to say, “This goes really good with this,” right? Like, we can cross-sell, we can upsell. The contact center not only has direct revenue implication because of all the data—that’s the insights to make better decisions for all the departments—but there is a sales activity, a very strong sales activity, that is happening there. Tell me about how your tool, or AI tools in general, can help support the sales activity, direct connection to revenue inside of a contact center.

[12:47] Ilai Szpiezak: Completely. That is such a good question. The way I think about this is that we take this process of gathering data, centralizing the data, analyzing data, sharing insights, and then taking action on the insight. This five-step process where I see what we are building, sometimes I think the decision making needs to still stay in the hands of a human, so I’m going to keep that out of this equation. But when it comes to gathering, analyzing, centralizing, and sharing, that is where AI has a lot more power than a human, because it can take the job to be done by someone and give the answers to someone to do something about then, or to share it and influence someone about it. That’s also interesting. So if we take what you said about contact centers being seen as a pre-sales and post-sales activity, then there is a lot of data that comes from pre-sales of what blockers we are facing that customers are saying, or hesitations, or the amount of the same type of questions that come from the customers. Are they talking about a competitor in comparison to us? Are they talking about a specific feature that we don’t have that they maybe would like to have? And this also happens a lot in sales calls—like, it’s also data, or should be data, to the company. 

I think that the time of growth at all costs—thinking about how we leverage these data points of the things that we are missing or the things that we’re doing well that we should keep doing because that’s why customers are converting or they are sticking on a post-sales perspective. And most of the answers to these questions that we are now giving as examples—so what we’re missing, what we’re doing well, what is working, what should be changed, what others have that we don’t have, and hesitations with pricing—all these different examples are things that we could answer, or we should be able to answer, through the data that we gather from the customers. Now, until today, it required someone to manually listen to these calls or go through each ticket and go in a spreadsheet and say, “This is X, this is Y. Oh, someone else talked about X, now we have two.” And the value of these customers talking about ROI is equal X amount of money. So if we take all this quantity of customers that are all in this bubble of money talking about the same feature, I can put a value to that feature, and I can go to my product team and say, “Guys, we have this amount of money on the line, we should do something about it,” and that goes for pre- and post-sales, I believe. So I do believe that there is a great direct connection from the customer insights we gather from the customer into the product being delivering to those customers.

[15:29] Steve MacDonald: So here’s what I got to do. I got to take the last two and a half minutes of what you said, provide it to every contact center leader that’s watching this, so they can share with their leaders, right? They can share it with the C-suite executives. Talk to me about the role of the executives in the CX leaders—like, just leadership’s role in making all this happen in a reality.

[15:51] Ilai Szpiezak: Maybe I will get told off by some people in the world, but I’m a founder, so it’s okay, and I will keep pushing the ideas for the future. But I believe that a leader’s responsibility is to service the customer first and foremost; they are leading a function that is supposed to service the customer, no question about it. I’m not saying they should stop. However, their very important task is to be the eyes and ears and voice of the customer to the organization. I’m not sure if that is in every job description or if it’s implicit or if we even think about it, but today there should be no question of why someone shouldn’t answer these questions when we are the leaders of what—we are talking to our customers. 

There is no other function in an organization that has as much customer insights as the customer-facing teams. So the job of that leader is not just to keep an excellent customer satisfaction score, and to have a great resolution time, and keep, you know, time to action and escalations, etc., all these metrics, stably and healthily, but their job is also to relate and share and answer the questions that these other personas need to action so that domino effect—we can keep the satisfaction high, we can keep the retention high, we can keep the churn low.

[17:13] Steve MacDonald: So I’m going to add something to what you just said, and then I’m going to ask you for your opinion on it. Because you’ve mentioned a number of times that there’s nothing more important than the voice of the customer that we need to be representing, right? And the way I started thinking about this is what you enable is for the contact center agents to the executives leading this to be the voice of a customer as a service to every other department in the company. 

If you think about it that way, your job is absolutely CSAT, NPS, keeping the customer, but you have a second really important job: to take all that learning and provide it as a service to everybody else. We are service organizations, all of us, to our clients; we just have to think the same way. As contact center executives, we eat, live, and breathe about customer service to our external customers. We don’t think that way to our internal customers. Why? We gotta ask ourselves that, right? This idea of, like, I think you enable your clients to be the voice of the customer as a service. How do you think about that?

[18:25] Ilai Szpiezak: I completely agree. I think that sometimes it’s because contact center organizations also have many, many years of maybe thinking about from in to out, from inside, we need to service the outside—the customers. They pay the salaries, we need to keep them happy, we need to make sure that they stick around. But because of that, we need more data to be able to do that job better. And that goes also to be able to have the data that we need at hand to answer the right questions. 

Today, I don’t even touch on that, because it’s not the area that I’m working with in AI, but there’s so many things that we can now leverage with new technology that is actually helping us to change the way we think about contact centers and customer-facing teams. Like you said, I love what you said, servicing also the other side, which is the internal customers, let’s call it. So not everybody in the team, I think, should be allocated to do both, because sometimes time is also not limitless, and we have a certain amount of time, and we need to do as much as we can with the customers. But as new technology helps us to have more time in our hands while providing a great customer service, we can take advantage of that time to better understand the data and to better make recommendations that help us then deliver a better service. 

I give another example that is not necessarily with support and service, but it’s still in the realm of customer-facing: a customer success team needs to know what is the status of what the customer asks for from the company perspective. Is this going to be released, and when or not? Should I tell the customer this is on the roadmap and it’s coming soon? Actually, it also applies to support and, you know, contact centers because we need to have visibility on this. Again, data, and insights. So, we need to make the connection between what the company is building and the customer is saying. The connector is the contact center. So that’s it.

[20:31] Steve MacDonald: So we’ve talked about so many different benefits in so many different ways, from revenue, sales, right on down. You started out by talking about your experience in CX, in customer experience, and in the physical world, like at events. And now CX is one of the most important differentiators in any business. McKinsey did a big study of 400 different B2C and B2B companies that were highly profitable, and one of their biggest conclusions was the biggest differentiator in all those companies was CX. How does what you do impact CX completely?

[21:15] Ilai Szpiezak: I’ll say a small story before that. That’s something that I remember while you were talking about this. And one of the greatest founders and CEOs, I believe, today is one of the founders of Airbnb, Brian Chesky, and one of the things he says is that when you build products only thinking about A/B testing, you stop building with empathy. And I think that customer experience is directly connected to listening to your customers and using empathy to decipher—not doing what they say, but deciphering what they need. And I think that’s also a job that is very, very hard to do and requires a lot of focus and knowing your customers. I don’t believe you should just take what they say and do it. I think we should understand what is happening and then decide what we do. But that just reminded me of when you were talking about that CX experience as a differentiator. 

So, I think it really connects to that. At the end of the day, when someone contacts a company, they don’t care that they’re contacting support; they’re contacting the company. Yeah. They don’t see, “I’m now talking to this person, so then that person is different.” No, it’s a company, right? When I contact Airbnb because I need help going to a new destination, I’m talking with the company itself. That first layer of contact is the company—it is the face of the company. What I’m building is—it impacts this CX experience and metrics and everything that the customer gets. 

But if we are building better products for someone, and if we understand who that someone is—whether it’s a B2B with different types of companies or whether it’s a B2C with large number of individual customers—when we understand what that end customer needs and why they need it, we can better build products that not only will make them happier, but will hopefully deliver a better experience, will deliver better information that they need to use the product. So it can also impact knowledge base articles; it can impact the way that they want to communicate with the company. 

It’s not just feedback about the actual product; it’s about the service as a whole. And if we are able to decipher the most important information right now, and give this right information with real quantities, which is what we do through Dolphin, to quantify this feedback to someone, there is no reason why that someone shouldn’t execute on that, because there is not just gut feeling—which connects to storytelling and empathy—but actual data-driven decisions that we all know that impacts the experience of the user on that CX. So that’s how I see it. It’s a little bit like a flywheel, where it’s a loop that should never end. The customer uses the product, will share with us what works or doesn’t work, we will take the data and make our decisions, and it will happen again. So that’s the way I see it.

[24:01] Steve MacDonald: I love it. In fact, I wish we had about another hour, but we’ve covered so much territory. There’s one last question that I really, really want to ask you. And that is that if there were just one takeaway from everything we’ve been talking about here, what’s the most important thing that you would want the audience for this podcast to remember after consuming it?

[24:25] Ilai Szpiezak: That’s an excellent question. I’ll answer it in two sets. One, whether they use a tool or not to understand customer feedback, they should start to think of storing this data in a way that, now or in the future, can be analyzed and can be actioned. Companies must think about what they’re doing today can very much be the compass of what the company does tomorrow. So think about the data that they are getting with conversations today and start to think of a process where these can be saved. It doesn’t need to be a whole flashy process with Dolphin AI and using all of these great—if they need, yes—but the starting point is to start storing it somewhere that someone can have access to later on. And the second answer to that, and going back to the title of what we are doing today, is that we need to know what we don’t know and everything we need to know. So essentially, have also an open mind that the data can inform what we need to know and not always what we want to know.

[25:34] Steve MacDonald: That’s really good, because sometimes what we learn, we don’t necessarily want to know, right? It’s like, “Shoot, we messed up, nobody really cared about that feature,” or “That launch just confused the hell out of everybody.” We need that. So that’s a really good takeaway. Here’s what I know: people are going to have questions. I know they’re going to have questions after this. Is it appropriate that we give out a link to your LinkedIn profile that people could reach out to you and ask questions?

[26:03] Ilai Szpiezak: For sure. I’m very active, and I’m vocal, and I believe founders should be. I’m active on LinkedIn, and it’s my name and surname. And of course, the company is Dolphin AI. The website is getdolphin.ai, and I am thrilled to share more about what we are building. I’m incredibly passionate about the new future of customer-facing and CX and support and service with AI, and the other side of the coin, is insights from the customer. And that’s what we’re building for a better insights repository, where anybody can access these insights whenever needed to build better products. So yeah, 100%.

[26:41] Steve MacDonald: Well, fantastic. Thank you, Ilai, for sharing all of this. Much, much appreciated.

[26:47] Ilai Szpiezak: Thank you. Thank you so much. It was great, great fun. Thank you, Steve.