Podcast appearance
Unlocking AI's Potential in Review Management
Why the public response should serve the next customer, not relitigate the last interaction.
Why the public response should serve the next customer, not relitigate the last interaction.
Nicole Alicea immediately reframes the audience for a review response: the original reviewer matters, but thousands of future customers may judge the business by what it says in public.
“The response that you're leaving is actually for the thousands of customers between now until the end of time who are going to be reading the way that you are responding to that negative review.”
Terminology note: this conversation uses the earlier word “facts.” RightResponse now calls the verified business knowledge used in a response “messages.” The transcript preserves the language used in the original recording.
A public response is durable marketing content. The episode examines how relevant business messages, negative-review judgment, competitor evidence, and structured AI steps create responses that help future readers rather than simply close an administrative task.
The response should address the reviewer appropriately, but it also needs to inform the many prospective customers who may read it later.
It determines the response objective and selects relevant approved business messages before writing, creating substantive differences rather than cosmetic wording changes.
It should acknowledge the experience without inventing a cause or making an unsupported admission, protect sensitive information, and use an approved contact path when an offline next step is appropriate.
Competitor analysis is a separate review-intelligence capability that helps businesses identify local strengths, weaknesses, and differentiation opportunities.
The original Create Brand NV episode and YouTube video are linked from this page.
Explore the Intelligent Review Responder
This transcript was generated from the original episode audio and lightly formatted for readability. Minor speaker or wording errors may remain.
[00:00:00] Coming to you from the Sunshine State, this is Create Brand Envy, a podcast dedicated to entrepreneurs and business owners, discussing businesses, marketing, leadership, and best practices in this ever-changing business landscape. Every week, we'll introduce you to a different business leader that has taken their company to new heights despite the odds, learn, engage, and thrive. This is Create Brand Envy. And now your host, President and CEO of Brand Envy, Nicole Alicea. Welcome to another episode of the Brand Envy podcast. Today's guest is George Swetlitz, co-founder of RightResponse AI. RightResponse AI is a review response and sentiment platform that's helping businesses use artificial intelligence to more effectively win customers. Now, what I love about George is that he's not your typical SaaS founder. He scaled a 220 location business, was President at a division at Sarah Lee Corporation, and he also holds an MBA from Harvard. So he knows a thing or two about running a successful company. Now he's built software that helps a location-based businesses with AI-powered review services without losing the human touch. So this is a tactical conversation with a big upside. I'm so excited to have you. We're going to learn so much from you today. Thank you, George, for being with us. Thank you, Nicole. It's great to be here.
[00:01:36] One of the reasons why I was very excited to chat with you is because I'm like, oh, maybe we can talk shop and convince other people. I'm a big champion. Believe it or not, a lot of companies are massively underutilizing their Google business listing. They're being business listing. There are so many ways that you can optimize it to attract the right customer to you. One of the biggest and best benefits of these location listings is that they allow anybody, former customers, to leave reviews on your business. I know I do not make any major purchase or do any kind of major decision without consulting Mr. Google and reading responses. I'm an expert at sorting them by the newest response, the lowest response. I think I literally read it in that order. Do you do the same? Do you always checking responses? Everybody checks responses. More people today read reviews than visit websites. If you're not bringing your website to your essentially your review responses, you're leaving a lot of opportunity on the table. We have, I ran into clients that are like, oh, yeah, you know, well, you know, we work really hard to get reviews and people leave some reviews and we're responding to it. You know, and then when I go check, I see some responses that are like, oh, my God, I cannot believe this guy, this client said that. So I speak to my client. I say, it's hard because it's like Mr. client, your review was not appropriate. And what I try to educate my clients on is you're not actually responding to that person that you had a negative interaction with your response, the response that you're leaving is actually for the thousands of customers between now until the end of time, who are going to be reading the way that you are responding to that negative review.
[00:03:38] And that in there in lies, you got to really separate that emotional aspect of what really happened with the account. Maybe the person is unhinged. It doesn't matter. You have to measure your response, considering other people that don't really know that this person is truly unhinged, you're going to read it and are going to form opinions about whether or not they want to do business with you based on how you handled that response. But the problem is that when you have multiple locations, it's a challenge for businesses to be responding, responding, providing responses that are on brand. So you came up with this idea of software. Tell me about that transition, like that realization. Sure. So as you mentioned, I was the CEO of a roll up of audiology clinics, we built the business to about 220 locations. And so in that process, we struggled with all these of these issues. How do you leverage reviews? How do you get them? How do you analyze them? How do you respond to them at scale? And scale can be different things to different people. If you're, you know, if you're an owner of a 15 location business, that scale, you only have so many resources and 15s a lot. But that same problem gets more complicated as you get bigger. And we had that problem and we never really were able to solve it.
[00:05:10] And so, you see, at the time, didn't really do it in the way that we were looking for. And very expensive. The tools that are out there are just phenomenally expensive. And when you have 220 locations, and you take 220 times, you know, $200 a month, you're talking $40,000 a month, it's just not. It was not something that we could do. But sourcing, you know, reading and responding to those reviews, or did you have like a full time, because they imagined it's a full time job, a full time person. Well, so we had 220 locations. We had people running those locations. Some of them enjoyed doing that, and they would do it themselves. Some of them didn't, and they wouldn't. And so we then had to step in as corporate and do that. And how do you do that? Well, we had a contact center. So when the contact center wasn't busy talking to people, they would open their computers and respond to reviews. But they would use templates and they would make things up. And it wouldn't sound like an owner. It was difficult to scale that. And that was the problem.
[00:06:20] And so what happened was we ended up exiting the business. I was sitting at home, and Jeff GPD came out. And I started thinking, you know, this, this technology, this disruptive technology, maybe we can apply this to this problem. And so it just so happened that 15 years ago, when I was a consultant, I built a technology team in India. And that team kind of worked with me wherever I went. And so I called them up and said, you want to tackle this? You want to start a company and tackle this problem. And so we decided to do that. And that's the origin story. Build essentially building the tool that we wish we would have had when we were all at this company. That is amazing. Okay. So first of all, is this like just one of those come, is it an organization, a company in India that does, you know, there's a bunch of software engineers or. Exactly. We had, we had started 15 years ago, a small software development company. And so when I would, you know, technology is such a software enablement is so important in any organization.
[00:07:41] And so whether I was acting on a consulting basis or in a company, I would leverage that team that I built in India, wherever I was. When I went to this business called Alpaca, the audiology business, they came with us and enable technology had Alpaca. And now we're working together to build this technology, this AI from the crowned up technology, focused on that review ecosystem. So what did you direct your, like what directives did you give this team on programming or code, I don't know all the correct terms, but like I guess training, how did you construct because we all know what chat GPT is and you can use it for a million thousand things. So it's flaws. So how did you turn around to the team and say, code it like this so it doesn't mess up because I love chat GPT, but by God it messes up, how was that? It was a real learning experience. It was an iterative process. We didn't know what to do and how to handle chat GPT and the other large language models that are out there.
[00:08:58] And we just had to learn through experience over time. And I think one of the things that we learned was that the more you try to do it one time, the higher the probability your effort will fail. And so we've gone in the other direction, say that again, say that again, and because like that sounded profoundly like deep and important, but that it just kind of like. So give me an example, I'm an example. Yeah, the more you try to do it one go, you know, when you're working on something and you, you try to get it to write a big article. It just, you know, it'll just give you the worst output. But if you say help me develop this one paragraph, yeah, it gets worse over time. But if you say I really want to work on this one paragraph, that's right. It's really good. Yes.
[00:09:51] So the smaller the ask, the better the answer. Yeah, and you have to drive it. You have to kind of like cons, I think I feel like I'm constantly like bracketing into, you know, to drive it. So sometimes I've been like, when I write something a really good paragraph with AI, do I own it? Or is it, or is it wrong? Because AI helped me so much. Like when I sit back and I think for myself, I go, I drove, okay, you know, I drove the truck. If somebody else tried to compose this paragraph by asking a prompt, it would not be the same output. Therefore, I really believe that even though AI helped me structure the sentence to be more clear, it's, it's still I'm the driver. I drove it there. Right. No, absolutely. And I think that's, you know, what you just hit on is, is our differentiation. So essentially, people talk about how, and I hear this from people about how AI is dehumanizing interaction and how it's stripping personalization out of interactions.
[00:11:02] What we've done is essentially the opposite in the sense that we create an environment where business owners can put information about their brand in our system. And what the AI does, which it's great at, is it identifies when that information is irrelevant. And when it's irrelevant, it incorporates it into the response. And what is it like an emotion? Is it positive negative sentiment? What qualifies as relevant? I'll give you an example. So thinking about about our business. If somebody wrote a review, and that review said, I had a hard time finding parking. There was no parking, and I had a hard time walking to the building. Well, you know, people, most people who are hearing is older. And so if they read that review, they would say, I'm not going to go there. I won't be able to get to the building. It's too long to walk. But if we were able to put into the system, if somebody talks about the fact that they couldn't find parking, remind them that there's entrance to the building, the parking lot for the building is behind off a beach street.
[00:12:12] And we have accessible parking and 45 spots spots. So the AI recognizes that that's relevant, incorporates that information. And essentially personalizes that response. The response isn't just saying, oh, I'm sorry, you had difficult finding parking. The response says, I'm sorry, you had difficult finding parking. But next time, go off a beach street, plenty of parking there. And so it allows you to extend. It allows you to say to customers what you would say to them if they called you and picked up the phone. Right. But they don't do nobody does that anymore. Well, I think it empowers a C suite minded person to program and train and state, like, you know, that response chain of reaction, like what to you is common sense that you tried, like, you know, the call center team to do, but they just couldn't, you know, they were kind of go rogue and they were or they would forget or you know that human error, you're really training and programming. So do you have, um, do you have like an accuracy rate that you reached before you said I'm ready to take this to market.
[00:13:23] Again, and that's where it gets back to the smaller, you know, the smaller the scope that you're asking, AI, the better. So essentially, if we're talking about the relevance of a statement, we would look at that and say, you know, we want that to hit 95% of the time. At the end of the day, AI doesn't know anything. It's just about the probability of words next to each other. That's all it is. And so it will never be 100% repeatable ever because the probabilities shift. And so you can't look for perfection. You can only look for kind of a high level of, of, you know, of getting it right. So, you know, out of 10 times that something comes up with parking, it might find that relevant nine times. It'll miss it once, but not the end of the world. And so that's the nice thing about the review ecosystem is that it's not medicine. Right. You're not trying to, you find a problem in somebody's body. You're trying to make a better review response. You're trying to make a better review request. You're trying to understand thousands of reviews better, but it'll never be 100%.
[00:14:46] So, for example, we do sentiment analysis. So the reviews come in. We analyze them. We try to do a lot of sophisticated things. So, for example, you know, reviews are crazy things. So, I have to McDonald's down the street. The burgers were terrible. So I went to Burger King. And at Burger King, I love the fries, but I hated the drinks. You know, I mean, there's so many things that are happening in a review. I've read reviews that are like an essay. And I'm like, I don't even know where to start. So, we break the review down. We look at every phrase. Is that phrase about the business? Is it positive? Is it negative? What topic? You know, we do all of these things in tiny, tiny chunks. But if you ran it through 10 times, would you get the same exact sentiments? No. But over a large enough set of reviews, you're going to get a really clear idea of what people are saying about your business. So, you bubble up like, is there any kind of reporting built into it where you bubble up sentiment or something kind of like, you know, this, you had this many positive reviews, this many negatives. How does your software work? How did you build the work?
[00:15:59] Yeah. So, in the area of sentiment analysis, there are lots of dashboards and you can drill down. It's really kind of neat what you can do. You can go in and you can see all the topics and the percent positive and negative for each topic. You can click on the bar and then you can read what all of those phrases were. And then you can click on a phrase and you can read the review that it was in, you know, it really allows you to understand how people are experiencing your business. And even more, if you want to, you can do that same thing to your competitors. Oh, so in your software, I can enter my competitors and get a report of the sentiment. 100%. That's fantastic. You can analyze your competitors using the same scorecard that you analyze yourself.
[00:16:51] Incredibly powerful. So, you are the man that has obviously studied and dissected the anatomy of the perfect response. So, what is the anatomy of an ideal response to any review, positive or negative? I think that the most important thing about responses. And actually, it goes back to what you said before that you don't write the response for the reviewer. You write it for all the prospective customers that are bottom of the funnel, trying to decide who they're going to call next. That's who you're writing it for. And if you put yourself in their shoes, the most important thing about that response is that it's helpful and informative.
[00:17:40] If you can do those two things, you will connect with those prospective customers. If they read a review, positive, negative, and you're not participating in the conversation by having a helpful and informative response, you lower the probability that you'll connect with that customer. But if that customer reads the response and says, wow, like, I now understand more about that business because I read the response. I'm going to say that business cares. I might give them a shot next. As a human, I have sometimes difficulty figuring out how I'm going to be helpful when someone's like really upset about something that kind of, you know, hey, I went to this whatever and to the store. And the receptionists were rude and didn't, you know, and dismissed me and told me to wait. And then they never came back. And I'm just really upset.
[00:18:40] I can, I as a human responder, and even less, a computer, be helpful in a, in situ, crazy situations like that where it's kind of like, sorry, you had a bad experience, you know. Yeah, I look at negative reviews. I kind of put them in two buckets, one or negative reviews, where you really can't help them, like the examples, they just had a bad experience. And I can't do anything about that. And so you just want to be empathetic, offer to speak with them personally. So, for example, in our system, if it's a negative review, and we assess that a conversation might be warranted, then we say, then it automatically will put contact information in and ask the person to contact the business. But there are other kind types of negative reviews where you can, the example about the parking, or if it's a restaurant and somebody complains about, we went there and it was noisy and we didn't expect it to be so noisy, we didn't expect there to be a band. Well, we can tell them where the schedule for the bands are. So next time they can check before they come. Gotcha.
[00:19:52] Or if somebody came for a walk on appointment and had to wait, we can say, well, next time before you come, call this number, and we'll tell you if there's a line, or we can make an appointment for you. So there's, there's different kinds of negative reviews, and we treat them differently. What does this offer do when it has no, when it's coming across a scenario that you never programmed it for? Does it kind of bubble up stuff that it can't deal with, or does it just automatically give a response? When there's no other information, it's programmed to kind of reflect back at a high level. So, you know, the typical AI responder, if someone says, oh, I, you know, I didn't like the burger. It'll say, oh, I'm sorry, I didn't help you didn't like the burger, right? It'll just parrot back what's in the review.
[00:20:44] And so that's not always, it's kind of like a dead giveaway. And so we, we program our system to operate at a slightly higher level. So if somebody just says something where you don't have a fact that can influence it, it responds by saying, we're sorry you didn't enjoy your experience with us, right? Because that's kind of what you would say to the very high level. It's very high level. It's not getting into the weeds, right? So it can't be wrong. It can't be wrong.
[00:21:16] And it's kind of what you would say if they were in front of you, hey, I'm sorry, is there anything we could do? Yeah, to get you back. Yeah, you don't know how much copy I read and I'm like, nope, nope, that that can backfire. It's too granular. Like, let's come up and use a more like a broader umbrella term where, you know, because we don't really know what's going on. So this words better to kind of cover this wide range of issues. How did you get so soon? I understand the programming the chat GPT part. How did you get chat GPT to talk to these response?
[00:21:55] So I'm imagining you're doing, you know, Google, my business being, how did you get them to kind of connect and, you know, respond on behalf of a human inside of that platform? Yeah, so, so for example, Google has a great API. Okay. And so we, we pull the data from Google and then we push it back to Google. And so it's all within the system. It's all, and people are going to automate that. So for example, we have a lot of customers that if it's a review without text, right, just the rating. Or it's a review with text and it's a four or five star.
[00:22:38] They'll just generate the response and push it back into Google. And they'll read, they'll review manually review the ones that are kind of negative reviews that have text. And then they'll publish it. We integrate over a hundred review sources. And so, you know, wherever people have reviews, depending on the industry, they're in different places. We pull them in for Google and, you know, we can push them. But for the others, we generate the response and then the user will copy it back into that platform.
[00:23:16] User being the client. The client. Okay. We, we've handled our agency over the years has handled, you know, Google business listings and handled responses to clients in the medical industry, where there's huge Hibba concerns. And we, we've kind of figured out we've kind of gotten formulaic about it. You know, you kind of, after a while, you kind of crack the code. But we're always very careful because if you're, if you're not careful, just a, just a little extra word can like, you know, acknowledge.
[00:23:48] You know, how have you guys had any particular experiences with the healthcare industry where Hibba and acknowledging is a concern? Yeah. So one of these little agents that we have running around in our software is a PHI agent. A personal health information agent that looks, you know, if you're a medical business, it looks through the review. And it identifies words and phrases that it, that should not be in the response. And it ensures that those words and phrases don't go back into the response. What are, I'm just curious what some of those are.
[00:24:25] If you happen to know, sorry to put you on the spot. I mean, yeah, no, it's, it's, it's a very complicated thing. But in a review, because technically, even writing a response, I mean, at a very technical level, even writing a response is a violation. Yeah. Because by writing a response, you're acknowledging their patient. So we like to use, thank you for your kind words. Right? I mean, thank you for your kind words.
[00:24:54] I mean, is that different practices have different levels of tolerance around that? So we actually have like a lenient and a strict and the strict is, you know, nothing. Oh, nothing like not at all. Well, I mean, it'll respond, but it won't say anything. You know, well, it is very strict about what it says back. The lenient one, if somebody says, oh, I had an MRI.
[00:25:25] I had an MRI and everybody was great. Right. The lenient checker will say, we're happy that you were satisfied with your treatment. But wouldn't, I was going to say, but even your treatment and obviously with us, it's saying it without saying with us. So even that. So yes, that's very lenient. Okay.
[00:25:46] Okay. The, the strict one would say what to say anything about it. Wow. Is there a, is there a moderate one? Well, we have to, we have kind of a strict and I mean, even the strict one is responding. Right? And you know, technically, you shouldn't respond at all.
[00:26:06] Or it'd be like, oh, thank you for your five star review. That's it. Yeah. Right. Our, our strict one will say a little bit more than that, but it won't acknowledge any condition. It won't even generalize a treatment. The lenient one will generalize a treatment.
[00:26:26] So for example, in the hearing aid space, it wouldn't say hearing aids, but it might say your device. Right. It's not saying it's a hearing aid, but a device. So it's just. You know, from our perspective, it's the practice that has to tell us how much risk they're willing to take. But I've never in all my years heard of anyone, you know, being penalized for those kinds of responses. And where people get in trouble is where when the, when somebody says something negative, and they respond, and they're like right at like what we did and what we didn't do.
[00:27:11] And they're talking about very specific things. That's when people get in trouble. When the, when the practice says something that the customer didn't say. If they, I was very, you know, I was very dissatisfied with, you know, the way you handled my case and the doctor writes back. Well, you asked for, you know, you asked for an ultrasound, but we didn't think it was relevant. And that's why we didn't prescribe it when they say something that the patient didn't say. Oh, so they're disclosing, yeah, because they're disclosing something.
[00:27:45] Gotcha. And honestly, it's like, you have to be a moron to write something like that. Exactly. But it happens. It happens. They get in trouble. Yeah.
[00:27:54] Oh, I've read some hilarious. Like I was sitting here giggling a second ago. If you're wondering what's wrong with her, she's just randomly giggling. I was actually remembering a review that I read where, and I don't want to share too much, because I'm still in the circles of this business owner. I mean, he called his client unhinged and said, you know, never call us back. And we never want to do business with you again. And you're, you're crazy.
[00:28:21] I mean, it was just, I'm like reading it going, oh, my God. Like, you know, funny stuff. Anyways, all right. So speaking of that, you know, you would, I find it interesting just looking at you as a person that created this, this company. Um, when it comes to establishing value in, in right response, AI rights. Like, so let's pretend we move through the fact, you know, all these concerns about AI and, you know, going to make mistakes is separating human. Let's move through all, all that.
[00:28:55] Have you struggled in convincing potential new clients that they need to be responding to every single review? Are there still people out there that are like, they don't value the response component. It's a double question. I'll finish my sentence. You can kind of think of it. Because the way that I'm noticing you that you are, that you are branding your company or that you're the copy on your website. You wrote reviews aren't tasks.
[00:29:26] They're your conversion engine. So when I read that, I'm a marketer. I'm done with different reader. Like, don't my brain does not think the way that normal people think I'm seeing interesting. He believes that people believe that reviews are just tasks, but he is elevating the value of right response AI by calling it a conversion engine and convincing his potential clients that this. Opportunity leveraged correctly can, can become a conversion engine for your company and ultimately generate more revenue. So just strategically as a Harvard MBA, you know, season business owner, share with us your how, you know, how you're dealing with establishing value in right response AI.
[00:30:20] I think people get attracted to us because we are at the forefront of leveraging AI in this review ecosystem, whether it's review requests or review responses were further along than others. And so people read about that and they come to us because of that. So you have the first mover advantage, right, the first mover advantage. ChatGPT told me that if I was looking for AI and reviews to call you. And I'm sure you optimized for that. 100% yeah. What we struggle with is the fact that people are so busy and so pressed for time that they that they don't spend the time necessary not all in some cases to set the system up to really help them create a conversion engine.
[00:31:23] So your claims have to set the system up and set like the parameters and you guys probably have them fill out a questionnaire. So for example, these facts. I don't know what the facts are like I don't know that they have parking behind, you know, off of Beat Street, I don't know that. So they have to put that in the system. They know what people talk about. So for example, we do is we when a customer signs up with us, we and we download their reviews. We look at the last 1000 reviews.
[00:31:58] Okay, that's a good sample size. We generate draft facts. So we look at what are people writing about. So in that example, we would say, you know, if people are talking about they can't find parking, we would have a fact it would just be sitting in our system that says if somebody says something negative parking, tell them that. We yet there's lots of parking across the street. Now, we don't know where the parking is. The person has to go in and edit that fact to say it's not across the street.
[00:32:29] It's on the back of the Beat Street. We do as much as we can. But people know what their business is about. People know the brand values that you're trying to communicate through their website or whatever it is. And sometimes people are attracted by the AI, but they don't want to spend the time to set up the facts so that you can do its job forever. Well, hopefully they're working with an agency like us who, you know, we know they're brand DNA. We're helping them, you know, so that's something that they can either delegate to an agency or their marketing person.
[00:33:07] That's right. But a lot of people don't. A lot of people aren't working with a, especially the one to five location businesses. I was going to say, like, I would imagine multiple locations, absolutely, or they have in house people. They have in house people, the larger ones, they eat it up. They love it because it's, you know, it's. They have all this stuff anyway. Yeah, it's a pain.
[00:33:30] And now, you know, now they can put it in. But it's the smaller ones that struggle a little bit. And from an agency perspective, which are exactly what you're talking about. When we white label with agencies. We talked to the agencies about the fact that this is a real kind of value provider for them. Right. Because they can say that they're client. We will set up the system. We will, we know about your brand.
[00:33:56] We will create the facts. We will set this up in such a way that it really reflects the voice that we've been trying to create. And so agencies love it because it gives them the tool that they don't have elsewhere. You can't, there's, there's not that level of, you know, granularity in other platforms. And if you're, like you said, if you're managing like, you know, three to five, maybe even 10 locations, you can do that manually. It's still not, you know, cumbersome.
[00:34:26] But once you start, I mean, 220 is like, how are you convincing these clients that that it's a conversion engine? Like how are you doing that positioning? And why did you choose to go with that positioning? Well, we show them, we just, we actually, when we, we have a free trial. So people will sign up. And then later, they'll schedule a call.
[00:34:51] And so, you know, they're in the system. And we'll just show them, we'll build a fact based on a review that they have. We'll show them how it integrates. And we'll say, if you were a reader looking at this and you read this response, how would that make you feel? Right. And when they see it and they think about it that way, that gets them,
[00:35:15] that gets them, that's when they see the potential. Yes. And the same thing is true about review requests. Everyone gets so many review requests now. And they're not personalized. Well, imagine if you could personalize your review requests. You know a lot about your customers.
[00:35:35] You know, how long, you know, in some businesses, you know, whether they've been a repeat customer, you know, what project, you know, you know, a lot of things about them. And to the extent that you can weave that into the request, it connects with that customer. It gives them another, it gives them a, you know, a psychological reason to respond. It feels human.
[00:36:02] Because only a human, only an account wrap that, that, you know, reviewed the account goes, oh, I've noticed that this is your fifth order with us. We're, you know, we're so happy. But if you wouldn't mind leaving a review. And so you guys program like to send an email or how do you guys handle the review request aspect of it? Yeah. So, you know, it can be through an integration with the CRN system.
[00:36:28] It can be an upload, it can be just manual entry for small businesses. But it's the same thing. They, we build or they build. It's, you know, we have kind of a, you know, our agents running in the background, but they write what they want that request to integrate. Like what are these fields that are coming in and how should they be used? And then the system just generates these very highly personalized requests.
[00:36:55] They can include photos. I love it. You guys are like surfaces. You guys are coming up with rules. And, you know, like if then, you know, and then just, and then letting it run with the magic of AI. Right. Exactly.
[00:37:11] All right. So we briefly mentioned something that, you know, I said, I said, oh, you optimize it for chat GPT. And then you said, absolutely 100%. And then I thought, oh, I bet, I bet my listeners would love to hear more about that. Because a lot of required, well, we're already working on it for our clients, like proactively. But I've already kind of heard commentary from some clients are like, you know, why don't we make sure that we're showing up for chat GPT, you know, and they're in those answers.
[00:37:38] Because there's a whole, there's this big movement now of like zero click. Do you know about that at the zero click? Can you talk about that? Well, when you say zero click, you mean people going into like the Google. And they're just getting the answer from this, you know, the AI. Yes. And they're also, they're also as instead of I started doing it a long time ago, where I wasn't even going into Google anymore.
[00:38:00] I was just asking chat GPT questions. And, you know, so, so we know we're watching website stats and, you know, tank. And it's not that the content and there's not valuable. It's just that AI is really, you know, pulling in and aggregating the data. And so a lot of people are asking, you know, how can you optimize your website to show up for chat GPT when somebody says, I'm looking for a nearby office furniture dealer. I'm, you know, recommend whatever.
[00:38:33] Like how do you, how can I use, how can I leverage AI to handle my Google responses? How did you guys program, you know, to, to be able to show up for that? And if you don't know, you can tell me, don't know, I understand your programers. Yeah, you know, part of it is nobody really knows how these language models are, you know, are, are building their own algorithms around this. And, but a lot of it is around what people ask, you know, so, the question that you ask chat GPT is typically longer than the question that you ask Google. And if it's a, you know, you, you tell it more.
[00:39:17] And so it has more information to go on when it's synthesizing all the information around it. So we, you know, one of the things that we started doing is every single blog post has lots of questions. So it's written in a question format because that's what people are asking chat GPT. And since our specialty is really around AI, we, we do that extensively in the AI review space. And so we have more, you know, we just, we've built that competitive advantage in chat GPT in that area. Can you mention earlier that you were one of the first movers? Do you have competitors or similar software?
[00:40:06] Or are you still, is the landscape kind of barren? There's, there's, there's one company out in Europe called Mara Solutions. They're doing very similar things, building facts into responses and things like that. They're the only one that I've seen other than us that, that is kind of tackling this idea of relevancy. Of facts about the business and building it into responses. I think, you know, bird eyes, try, is starting to do that. Bird eyes, one of the big players in the space.
[00:40:41] Fabulously expensive. So, you know, in order to get these features, you have to spend a ton of money. But they're doing it by kind of this notion of uploading documents. You know, like a, a rag type system, you know, where they're uploading documents, and then they're letting the AI figure out what might be relevant. I just think that's so much harder. I think the probability of.
[00:41:11] I think people don't a lot, I hear sometimes people talk about chat GPT, and I'm, I'm just sitting, I don't say anything because I, I'm not entrenched enough about it to like, you know, educate anybody like, you know, or lecture them, but I do know enough to know that it does not understand. It is not, it does not probably, it's just, it's a, it's tokens and calculating the likelihood that one word is going to be next to the other and. Yes, so think about a review without text. Great example. When we first started, we didn't do anything special for reviews that were rating only.
[00:41:50] We called from one of our customers and he said, you know, you got it, you got to do better. Because all my responses are sounding the same. And so I went and I looked at it, he had a lot of reviews that were rating only. And it was saying the same thing. And so I start thinking I talk into my team and different people. And the problem is is that when, when AI has nothing to go on. It looks at the most likely set of words that should be next to each other.
[00:42:18] It's often going to be the same thing. Thank you for your kind words, right? Like you said, that's, that's out there so much that thank you for your kind words is a very common. So just as an aside, we have pro, we, we have a black list. And one of those black list items is thank you for your kind words. That's why you didn't say anything because you were like, you were like, yeah, that's on our no, no list. I'm not going to insult her. So.
[00:42:47] But what we do is we have a page where people can write in the kinds of things they like to say. Right. Different phrases that they like. And then the AI randomly selects those phrases. Along with all of the other customizations, how do you want to greet the customer? How do you want to close? To the business name, all of these things.
[00:43:15] And it creates a really nice response. And because it's random and using AI, no two responses sound alike. And so that's so we had to actually intervene. Yeah, and programming to kind of rotate, you know, off of these. That's right. But I mean, again, if you understand programming and coding, then those that's not hard. I'm thinking about your team in India.
[00:43:48] And, you know, I think everybody has tinkered a little bit with hiring somebody abroad. Do you have any best practices or any advice or any kind of feedback? Because clearly you've done it successfully for many years. Well, I mean, I got lucky. You know, I, I went through a couple of people when I for this is 15 years ago. That and then I found a guy and he was just a wonderful person. A great developer had the capability of building a team.
[00:44:22] And we never look back. So, you know, you got to find that one person. And because you're not there. And you have to rely on someone. And so you have to find that one person and then let them do their thing. And we, we just got lucky all those years ago. You know, one of the biggest challenges that I've had and that I've just seen across the board,
[00:44:46] like, you know, finding really good people to work with you. That's something that is, you know, a whole different subset of conversation. And so I've been in a lot of roles, including this one. And you said, luck. And I'm just kind of sitting back going, because there is that luck component. Because I can't tell you how many times I've tried to make interviewing, recruiting, hiring, formulaic, right? If this, then that or if you notice this, then that's what this means.
[00:45:15] And I, for the life of me, your way ahead of me, if there's any insight, please share. Because sometimes feel like it is just dumb luck that you end up with the right person or just unlucky, right? That you end up with the wrong person. I don't know. What are your thoughts on that? You know, I've hired a lot of people over the years. And I don't know that I have a formula.
[00:45:41] You know, I've hired people that I thought would be amazing and they were duds and vice versa. So, you know, I think you do the best, you do the best you can. You, you have to make sure the core skills are there. But the bigger part of its personality and fit. And you really don't understand that until they're there. And I think, you know, the same problem is on the other side. You know, candidates go through these excruciating interview processes.
[00:46:12] And then they get into the job and they realize they don't like the culture. And so, you know, when you find someone that fits, you really have to work at keeping them. Yeah. And it's comforting and frustrating at the same time to hear that that that that luck component. But it's true. And in business as well, you know, when you, I think set off as an entrepreneur, there's definitely a bit of luck involved. And, you know, whether you make it or not, there's certainly a lot of skill sets that you can apply.
[00:46:48] Yeah, a lot of its luck. It really is. You know, when we were in the, when I first got involved with the audiology business, COVID happened. Oh, wow. We had to shut down 220 clinics. And then reopen them.
[00:47:08] And so, you know, there's skill. And then there's the things that happen that you just can't plan for. And where the skill comes in is, how do you handle those, you know, unknown situations? And what are your best practices for walking into these unknown, uncontrollable situations? Data. Lots and lots and lots of data. So, for example, at this particular company, we decided to use the COVID shutdown as an opportunity to understand which clinics we didn't want to reopen.
[00:47:50] Right. So we went through a very careful analysis about the competitive markets and the financials and our prospects for success because why open. A clinic that we didn't feel. Right. So when we came out of COVID, we ended up with higher profitability because we had taken the opportunity to not reopen some clinics that were marginal that we didn't think we could fix. Yeah. And that the case that like every day, you're always kind of stuck, I think in the undertow of just the day to day operations and you don't actually ever get a chance to like take a breather and stop and actually look at exactly what you said, COVID finally gave you the chance to do.
[00:48:34] Which I think more people should carve out more time for that, you know, more leaders to really just take a step back and really look at. Exactly that, like what are the top performing products, the services that we have, what are the top performing locations that we have, like, you know, because it's, it's important work, but it always kind of gets like, OK, well, maybe, you know, later, later, later. I have another idea for you. When I, when I was, you know, this business, I would allocate four hours a week to just talk with anybody. So if I got an email and someone had an, you know, was pitching something. And I thought, you know, maybe there's something there. I would give that person 20 minutes.
[00:49:24] Call people, you know, in Europe that we're doing interesting things, I would email them and say, hey, I, I read about this that you're doing. I'd just like to talk to you. Maybe there's something, maybe there's not. And so, you know, you have to be out in the world, right? You have to be out. If you think you have all the answers, you will never build your business. You will never excel. You have to recognize it's a big world. There's a lot of people out there smarter than you.
[00:49:57] And your job is to find them and do better through other people. And so, like you said, you have to carve time out to do different things. Part of it is stepping back, thinking about the business. I found that stepping out and talking to people, doing different things. Sometimes I would approach them. Sometimes they would approach me, led to a lot of great opportunities. Agreed 100%. What a beautiful way to close on that note of wisdom.
[00:50:34] I certainly appreciate your time and insight sharing your journey of creating right response. I was just such an innovative and like you said, kind of disruptive idea. I admire your boldness in saying, I'm going to, you know, develop it, bring it to market and, you know, offer it. It's great that you guys white label with agencies, this agencies. We are sometimes like to say that we're the primary care doctor for the business on the marketing end where we're kind of, you know, looking at the entire organism. And then like, we also, we hire like, you know, specialists like, you know, so people that only do websites, people that only do graphic design. And I kind of, we kind of lasso them together for the benefit of the client.
[00:51:17] And we're also looking for white label partners or doesn't even have to be white label. Just partners that can help us bring more value to our clients. And a lot of work that we do is taking meetings with people like people like you and, you know, or any other vendor that approaches periodically will meet with them just to see what they have. And then if it's good, we bubble it up to the client. And so there's, you know, like you said, we do live in a very complex world. And we definitely need each other and we need to, you know, all this technology is great. It brings us together.
[00:51:48] In many ways, it also kind of like sets us apart. And I'm just so grateful for the time you shared with us. How can people get in touch with you, connect with you, follow you. So our website is right response AI dot com. And there's the ability to schedule a call. He's go right there and schedule call. And if you say I want to speak with George on the skull and the notes, then it'll be a meal on the call.
[00:52:17] Or people can just reach out to me directly. She sweatlets. I'm on the website on our team page. She sweatlets at right response AI dot com or through LinkedIn, either through right response or me personally. Perfect. Well, thank you so much, George. It was such a pleasure to have you on the great brand NB podcast. It was great to be here. It was a very, very interesting conversation. Thank you. You're welcome.
[00:52:43] Thanks for listening to create brand NB. Be sure to subscribe wherever you heard this podcast. Never miss a future episode. Brand NB is an integrated marketing and advertising agency that helps brands innovate while maintaining their focus and identity. To learn more or to get in touch with Nicole, visit create brand NB dot com. That's create brand and the letters NB dot com. We'll see you next time.