AI Receptionists:
Transforming Customer Service!

AI's Role in Customer Communication

Michael Krigsman: AI is reshaping customer communications with terms like virtual workforce and AI Receptionist. I'm Michael Krigsman, and I spoke with Damon Covey from GoTo to explore how small companies can use AI agents to improve both efficiency and customer service.

Damon Covey: One of the main customer service challenges is communicating effectively with your customer over the channel and how they want to talk to you. If a customer wants to text you, or if they want to reach out to you via Instagram or Facebook, or similar platforms, it's about being able to communicate with them that way. Traditionally, for customers, this meant they had to use all those different platforms to communicate, and they often lacked a unified way to do that.

Now, there are solutions that bring all these things together. This makes both the employee or business experience and the end-user experience as seamless as possible. As customers' needs have evolved, we've added more and more channels and opportunities to communicate over the years to that cloud communication solution.

We're in the business of providing the opportunity to talk to customers, communicate with them, schedule appointments, and do things of that nature in the way that customers want to talk with the company directly. That's a traditional challenge, especially for smaller customers.

Michael Krigsman: Can you give us a few examples?

Damon Covey: If you have a dental practice and someone wants to communicate with you and set an appointment, they are very often just going to be doing whatever they do during the day, and a text is much faster to set that appointment. That's one example.

Another example I can give you from my personal experience: I had a pipe that burst in my basement. I was trying to reach out and find a plumber. I didn't know a plumber, so I started calling down the list of plumbers, and the first one to text me back got my business. Being able to communicate with customers in the way they want to and the way they communicate in flow are critical parts of a small business practice.

Michael Krigsman: Making it easier for customers to interact with the business.

Damon Covey: Making it easier, and also having it integrated back into the systems they care about, their CRM or other systems that they might be working with, keeping them in those systems to make sure that they are tracking that information as well.

Michael Krigsman: The integration with their existing systems, their existing processes, is essential here.

Damon Covey: It's critical to make sure that they can continue to do business the way they want to, but also communicate with their customers in the way that customers want to communicate with them.

AI Virtual Workforce and Agentic AI

Michael Krigsman: How can AI make this process easier, better, faster?

Damon Covey: AI has the opportunity to really help. Especially if you look at a small business, they are most of the time multitasking. They are dealing with customers. They are also dealing with things inside of that business.

They don't always have the opportunity to answer the calls or answer key questions, things of that nature. They might actually lose some of those customers that are calling in.

AI has the ability to be a force multiplier and answer the phones 24/7 for them, as an example, answer specific questions, and then transfer that information along with the caller, when appropriate, to the right person who can help them and make that a more efficient process.

Michael Krigsman: Damon, I've heard you use the term virtual workforce. What is that?

Damon Covey: A virtual workforce is essentially a way to harness AI technology to do things that you would like a human to be able to do, but they may not be available. If you think about humans in your environment, they are not only answering calls and texts and things of that nature but also doing other things. You want to automate those repetitive tasks.

AI can take a lot of that over. It can start to automate that, whether it's answering the phone (in some cases we've talked about), or possibly answering questions about locations, hours, things of that nature. It's about really using AI in a very smart and practical way to automate those tasks. That's where you start to see the AI virtual workforce emerge, and we'll see more of that as we go into the future as well.

Michael Krigsman: We hear a lot about AI agents and agentic AI. Is a virtual workforce the same thing?

Damon Covey: It's a part of it. If you think about a virtual workforce and agents, that's the tip of the spear for answering calls, automating mundane tasks, and things of that nature. What agentic AI does is take that information and act on it.

If you need something to happen in a somewhat autonomous fashion, that's where agentic AI can come into play. For example, if you wanted to pull something out of a CRM and then insert that back as you are talking to a caller, that would be an example of agentic AI on the front end. It will get more sophisticated as we go forward and perform more sophisticated options and advanced actions as well.

Michael Krigsman: And how is that different or the same from the virtual workforce?

Damon Covey: It's part and parcel of the same thing. Virtual workforces will just mature over time. To be able, again, to automate those tip-of-the-spear tasks and also to automate other tasks that are smarter, that require connecting to multiple different systems and require multiple different pieces of input, that's where agentic AI really has power. Its power is in connecting those things together and automating a workflow to satisfy a particular problem that might dip in and out of five or six systems, as an example.

Michael Krigsman: And then how does the virtual workforce relate to that workflow that you just mentioned?

Damon Covey: The virtual workforce will be at the front end of that. An example of this is AI Receptionist, which is able to answer the call, understand the intent of the caller, answer some simple questions, and then potentially transfer that person over. In the future, we see that getting more and more sophisticated, where it can actually take actions as opposed to just answering the call. That's how the two fit together.

AI Receptionist and Its Capabilities

Michael Krigsman: You've just released an AI Receptionist. Tell us how that works.

Damon Covey: AI Receptionist is something that anybody can add into their telephony system that will answer phone calls for you. If you think about it, Michael, the most challenging part of AI as it relates to customer communications is the voice section. That's why we took that one on first.

What this allows you to do is place it at the front end, or frankly in the middle or even at the end, of your telephone call to answer that call, answer questions, and transfer that over to the appropriate person on your side with the intent also summarized.

Imagine, previously, you were just getting a call from someone out of the blue, and you didn't know what they were calling about. We can capture that information and transfer that over to someone so they can actually understand what this person is calling about: Are they upset or not? What are they actually looking for? You can automatically know how to approach the call and how to better handle that call.

Michael Krigsman: The virtual receptionist interacts with the caller and feeds information back upstream to a person who may or may not interact with that caller. Is this correct?

Damon Covey: We can also extend that AI Receptionist to answer, again, common questions. If you want to help that AI learn over time by feeding it information that it should answer as you start to observe what's happening, you can. Yes, ultimately, it probably ends in some human interaction, and the idea is to augment the human interaction with the caller and make that resolution as speedy as possible.

Michael Krigsman: To some degree, the virtual workforce is an autonomous process that interacts with the caller.

Damon Covey: Absolutely. On its own, with your permission, of course, it will answer those calls and answer those questions. But it can be, and it does operate, within its sphere of influence, autonomous from someone else actually having to manually take that call or take an action.

Michael Krigsman: And so there's a set of boundaries that are built in. You mentioned scope or sphere of action.

Damon Covey: Correct. That's a really important part of this, actually, Michael. What we don't want to do is just wire an AI Receptionist to an LLM. That would be a bad idea because it might actually give some bad information or something that you weren't expecting it to.

Making sure you tune the AI Receptionist to give only the proper information with the guardrails you want it to have, and starting with a very scoped set of responsibilities and building up from there, is the way to help it learn within your system.

Michael Krigsman: The set of boundaries is vitally important here.

Damon Covey: Absolutely.

Introduction to AI Receptionists and Their Benefits

Damon Covey: It's probably the most critical part of this to make sure, and again, what I always tell people is start with something practical. Start with something that's a simple problem, it's repeatable that you can solve over and over again, that you feel good about an AI Receptionist talking about, and then build up from there as you start to see how those reactions are going as you go forward.

Michael Krigsman: Damon, what benefits does a virtual workforce bring to GoTo's customers?

Damon Covey: It starts with really the ability to have any interaction handled 24 by 7. If you've got your employees that are doing something else and they can't answer the call, that's number one. The other thing is ensuring that you can free those employees up to do higher value things, more beneficial things, and look at other opportunities within, within your environment. If they're doing the same thing over and over again, now we can free them up to do that. And then I would say the other thing is really just making sure that because you're answering swiftly and giving them feedback right away, really enhancing customer satisfaction from the end user perspective. They're getting information they need faster than they would have previously.

Michael Krigsman: This notion of improving customer service for the caller is really important. Can you elaborate on that?

Damon Covey: With an AI Receptionist or AI, you have the ability to guarantee the response that the user, or the caller in this case, is getting. And that gives them the ability to:

• Automatically have their call answered at all times, that number one is, is a primary benefit.
• Get a guaranteed answer on what that is.
• And if they need to be, have a guaranteed way of getting interaction.

Those are all three things that customers are looking for, that immediate response. They want that immediate interaction. People don't wait anymore to get feedback from companies they want to do business with, and this gives them that immediate feedback.

Michael Krigsman: How is this different or more beneficial than traditional customer service channels?

Damon Covey: Traditional customer service channels always need a human intervention. And this is not meant to replace that human intervention, but what this does is helps you scale your business. And that's where it's special, is again, if you're a small customer, part of what you're doing is just, you know, the business of making sure that your business is done and scaling it, making sure you can answer all those calls, making sure those, that without having to have an agent behind every single call. And again, freeing up those agents to do more high interaction, better sales process, better service process, things of that nature. That's really what this is about is augmenting your human workforce so that they can do things that are higher value, again, to the company on the back end.

Michael Krigsman: There's an efficiency benefit to the GoTo customer and a service benefit to the end user that's calling in.

Damon Covey: Correct. You definitely get a benefit both ways in terms of accuracy and in terms of how fast the call is answered, in terms of how fast it's responded to, which will by itself gain you a higher satisfaction rate.

Deployment and Integration of AI Receptionists

Michael Krigsman: Damon, let's talk about deployment. What is involved with implementing or deploying this kind of system?

Damon Covey: That's a critical decision factor. Small customers could not, you know, capture something off the shelf and implement that on their own. The AI Receptionist as an example introduces the ability to have a small and simple operation that you can feed things like KBs, knowledge base articles as an example. We can feed FAQs into it, right? We can start to actually put questions in there, things of that nature.

It makes it very, very easy and practical to get up and going off the ground. And then I think the other thing that's really important about this, Michael, is you can observe the system over time as well. That's a critical differentiator also is it's, yes, it's easy to get up and running, but can I see how it's actually doing? Can I see how the customers are responding to it? Can I listen to the calls? All the above is there and so you can tune that as you go. Both of those things are critical factors in implementation.

Michael Krigsman: When you say see what the system is doing, can you elaborate on that?

Damon Covey: Yes. One of the most critical things with any communication system is, is you can run analytics and you can start to see things like a word cloud of: Was someone angry, right? Did someone, you know, say something that they shouldn't have, right? Are they happy? And digging into that and seeing what those interactions actually consisted of and listening to the call if you need to. That gives you the ability to run quality metrics against your environment, which usually is reserved for very large corporations. That is one of the biggest barriers to entry to having anybody, whether it's an agent or an AI interface, in this case, answer calls, is understand that quality behind the scenes.

Michael Krigsman: And because of the AI, I'm assuming that this turnaround is very fast.

Damon Covey: Almost immediate. The call is recorded if you want it to be, and then the call is brought forward and then we actually use AI to do a summary analysis of that, and almost immediately you will see what happened and you can drill into that and diagnose what occurred with that interaction.

Michael Krigsman: Damon, what's involved with deploying this kind of system?

Damon Covey: The key here is to making the system easy to deploy, and making it practical to deploy. What you would do is you would just start feeding in knowledge-based articles. It can crawl your website information. You can even put simple keywords in there to say, "If you hear this, then say this," or, "If you hear this, transfer this to an operator or transfer this to a salesperson." Really, it's very practical to set it up. It's very practical to have it learn with you, and only the things that you feed into it will it actually be able to respond back to.

Michael Krigsman: What about integration with existing systems and existing business processes?

Damon Covey: A couple of things. First off, it's very important to be able to integrate with your human employees. We talked about the ability to capture the sentiment of a call and make sure that gets transferred onto a human. That's an important part about this. But the other thing is the ability to integrate with your systems. If there is information we capture, being able to integrate that back, and that's one of the things you want to look for is how does it integrate with the system that you're mostly in, and usually that's a CRM.

Making sure you've got the proper CRM integration, making sure posting that proper integration into the system is a critical component to make sure that you don't have to go back and forth between systems as you're moving forward and working with your human employees.

Michael Krigsman: You've mentioned the data several times and the training of the system. Tell us a little bit more about what kinds of data are used.

Damon Covey: Simply put, any kind of data that is in a written format, in a digital written format, you can feed into the system. Again, FAQs, knowledge-based articles, those types of things. But the other thing is the notion of observability and the ability to go in and tweak and optimize the system. That's critical, right, to see what's happening with these interactions, see what's happening there, and then you can either tweak the documents accordingly, maybe put some additional prompts in there for that. But that's something that's, you know, part of… As an example for what we've done is, we built that into the interface. You can just feed those documents in. You can tweak those questions if you want to and all of those kind of things. That's all part of the data that goes in to make sure you can get a quality response back out.

Michael Krigsman: Do you have recommendations or best practices for folks that are just starting out?

Damon Covey: Start with a simple, practical issue you're trying to solve. If you're getting the same call for the same thing over and over and over again, that's very tangible, it's very understandable, and you can usually put the information you want an AI Receptionist to say back to the customer very, very quickly.

Again, in the case of GoTo, we've allowed you to come in there and put simple questions, simple answers. When you hear this, you can respond with that. And as you do that, you'll find that you can tune that very quickly to the other things you want to do. You'll gain reliability. You'll gain trust. You'll gain the ability to get more confidence and then train that AI Receptionist as you move forward. But starting with something very practical and something very simple is a great way to get started and makes it very real for businesses to get a handle on.

Michael Krigsman: Starting small enables that trust level to grow because it is new.

Damon Covey: I'll give you an example, Michael, of where we've seen something really work really well. I have a medical shop that I work with, and they have an after-hours answering service. And the after-hours answering service is literally writing down each person that calls, and then they're faxing that information in to the office the next day. Nothing wrong with that. That's happening today.

But the AI Receptionist can do the same thing. It can actually take those calls. It can capture what they called about. It can capture the person's name depending on what you want it to do. And that's a very practical thing you can start with that makes it easy for you to work on and tune and adjust as you go forward.

Collaboration Between Virtual and Human Workforces

Michael Krigsman: Can you talk about the collaboration between the virtual workforce and the people working at the company?

Damon Covey: It's absolutely critical to make sure that the AI workforce can communicate and can talk to and work with your human workforce. You might want to, in some cases, have the AI workforce at the beginning of a call. You might want to have them towards the end of a call. For example, if you wanted to ask for a post-call survey or something of that nature. But make sure that you can transfer the call and intent, automate those responses that you want as a business, and ensure that we're not replacing the human because the human touch is still needed. And I don't see that going away anytime soon.

But making sure you're augmenting what the human is doing by taking those common repetitive tasks out of their view so they can go do something more special… that's the kind of interaction, the kind of specialty and specialist that AI can bring to the market.

Michael Krigsman: You're really talking about redefining some of your processes to accommodate this new aspect, this new virtual team.

Damon Covey: Yes.

Optimizing Workflows with AI

Damon Covey: Great way to look at it. As you take those mundane tasks out of your workflow, what can those employees be doing that provides better value to the company, that provides higher quality interactions with the customer? Because those things will always be there. People will always be needed. But taking some of those things that, you know, in some cases can be frustrating because you're answering the same question over and over again, automating that, and then making sure you can work with a human at the end of the day, those are critical aspects to making sure that we have that human to AI interaction nailed.

Michael Krigsman: Do you have any advice on this notion of redefining the roles to work successfully with digital workers?

Damon Covey: What can you do to delegate administrative tasks, as an example, or things, again, that you don't want to do? One tangible example of this is if you take a call, at the end of the call, someone needs to summarize the notes into the CRM, as an example. And what AI can do for you very quickly is summarize those notes for you.

Now, you can, of course, go in and edit those if you want to. But having them be posted in for you automatically as a summary that you can accept or you can edit later, that is an example of an administrative task and an overhead that you probably don't want your employees to have to worry about. As they do that, then they can start to put their time and attention towards other things. I think it's critical to look at those, at those parts of it and then figure out what the opportunities are that maybe you haven't been able to look at before with your employees that they can now lean into.

Michael Krigsman: So again, you're really starting with simple labor-saving use cases.

Damon Covey: Correct. Exactly.

Addressing Privacy and Security in AI Systems

Michael Krigsman: What about privacy and security concerns?

Damon Covey: First of all, make sure that you're using cloud systems and AI systems that have published security policies with encryption, fail-safes, things of that nature. Can't say enough about that and making sure that those things are happening. Usually, that's happening with a commercial vendor, but you want to absolutely double-check that that's the case. You also want to make sure that you tune the AI to respect your privacy policy.

If there are things you don't want the AI to ask, for example, if you're medical, you probably don't want to ask for certain patient information. You don't want to ask for things like Social Security numbers. Tuning your security policies with the AI to make sure that it doesn't ask for things that you don't want to get access to, make sure that not only is the technology safe for you to use, but you're keeping your customers and your business safe from anything else that might come in and be captured accidentally as part of the system.

You protect the company and then also the users because there's a natural curiosity about how AI will handle our data and our privacy. And you can secure that by making sure that your customers understand it and you're transparent with them about what you're capturing and what you're not through your AI systems, which is a critical component of success.

Future of AI in Personalization and Action

Michael Krigsman: Damon, where is all this headed?

Damon Covey: We will see much better personalization and the ability to interact with people as if they were talking to someone. Imagine, as an example, if you called into, like, let's say, an automotive dealer, as an example, and I know you own a 2024 Toyota 4Runner. I could bring that information in and ask and say, "Are you calling about that particular vehicle?" Or I can call, I can ask about something that's very, very personal towards you. And I think that's one thing you're going to see. The latency will improve. The ability for you and I to talk and an AI to talk exactly like we are at the same pace will be very critical.

And then I think the other thing you're going to see is we talked a little bit about agentic AI and the ability to take actions on your behalf. And as tools emerge and become more integrated with your different systems, being able to take more actions on your behalf and make that more seamless, and that's really where you'll see a force multiplication for businesses of where we can take, it can take common actions, not just answer common questions, but take common actions that might span different systems in your environment.

Those are things that are happening. Many of those are there today. It's only going to get better as we move forward.

Michael Krigsman: Exciting times ahead.

Damon Covey: Absolutely. Making it practical for all customers to be able to adopt is what our mission's all about.

Michael Krigsman: Damon Covey from GoTo. Thanks for taking time to chat with us today.

Damon Covey: Thank you, Michael. Great to be here.