Reading Time: 33 minutes
Ever wished you could turn every customer interaction into a personalized, seamless conversation that boosts your sales and keeps your customers coming back?
Many of us are stuck with outdated methods that don’t scale, don’t deliver, and leave customers frustrated. Meanwhile, we’re missing out on huge opportunities.
It’s frustrating, right? You want to provide an exceptional customer experience and see a return on your investment, but it’s hard to keep up with the demands and expectations.
Enter Kelly Noble Mirabella. She’s a digital marketing guru and the brains behind Stellar Media Marketing. Kelly has cracked the code by using AI to make smarter, super-effective chatbots that totally transform customer interactions.
With over a decade in social media marketing, Kelly’s not only a sought-after speaker and consultant but also the COO of Chatbot Builder AI. She’s here to spill the secrets on how these AI chatbots are game-changers for boosting engagement and driving impressive ROI. Get ready to rethink how you interact with your customers!
LISTEN TO AI IN MARKETING: UNPACKED:
WATCH AI IN MARKETING: UNPACKED:
AI in Marketing: Unpacked host Mike Allton asked Kelly Mirabella about:
✨ Smarter Engagement: Understand how AI chatbots provide more intelligent and personalized customer interactions.
✨ Boosted ROI: Learn how integrating AI chatbots can lead to higher lead generation and return on investment.
✨ Getting Started: Gain practical tips for implementing AI chatbots, even on a smaller budget.
Learn more about Kelly Mirabella
Resources & Brands mentioned in this episode
Full Transcript
(lightly edited)
How An AI Chatbot Is Redefining Customer Interaction and Boosting ROI with Kelly Mirabella
[00:00:00] Kelly Mirabella: I think anytime that we’re using AI, we need to be very clear that it is AI. I think a lot of times business will say, Oh, I don’t want to tell him it’s an AI, you know, customers won’t like it. I’m going to tell you right now, customers don’t care as long as they get their answers. Like, all the data points to this.
They would rather the chatbot just fix the problem than have to wait and talk to a human. So we need to be very straightforward with setting that expectation that this is an AI chatbot. The other thing that whenever you’re building chatbots, You have to make it easy for them to opt out. Other thing I would say, you know, to the, those people who are like, well, if it’s AI, they’re not going to want to talk to it, it’s going to chase people away. In my experience with a well built AI chatbot, people forget very quickly that they’re chatting with an AI chatbot. We’ve actually had customers reach out to us and we have like reviews that are saying, this blew my mind that this was AI, like, I am so surprised. I actually forgot.
[00:00:54] Mike Allton: Welcome to AI in Marketing: Unpacked, where we simplify AI for impactful marketing. I’m your host, Mike Allton here to guide you through the world of artificial intelligence and its transformative impact on marketing strategies. Each episode will break down AI concepts into manageable insights and explore practical applications that can supercharge your marketing efforts.
Whether you’re an experienced marketer just starting to explore the potential of AI, this podcast will equip you with the knowledge and tools you need to succeed. So tune in and let’s unlock the power of AI together.
Greetings program. Welcome back to AI in Marketing: Unpacked where I selfishly used this time to pick the brains of experts at keeping up with and integrating or layering artificial intelligence into social media, content, advertising, search, and other areas of digital marketing. And you get to learn to subscribe to be shown how to prepare yourself and your brand for this AI revolution and come out ahead.
Now, have you ever wished you could turn every customer interaction into a personalized, seamless conversation that boosts your sales and keeps your customers coming back? Many of us are stuck with outdated methods that don’t scale, don’t deliver and leave customers frustrated. Meanwhile, we’re missing out on huge opportunities.
That is frustrating, isn’t it? You want to provide an exceptional customer experience and see a return on your investment, but it’s hard to keep up with the demands and expectations. Enter. She’s a digital marketing guru and the brains behind stellar media marketing. Kelly has cracked the code by using AI to make smarter, super effective chatbots that totally transform customer interactions.
With over a decade in social media marketing, Kelly is not only a sought after speaker and consultant, but also the COO of chatbot builder AI. She’s here to spill the secrets on how these AI chatbots are getting started game changers for boosting engagement and driving impressive ROI. Get ready to rethink how you interact with your customers.
Hey, Kelly, welcome to the show.
[00:02:49] Kelly Mirabella: Thank you. Wow. Can you just introduce me for everything I do going forward? You are fantastic.[00:02:57] Mike Allton: I have been taught by some of the best, but the best part is I’m just sharing what you folks do. I always tell my guests that I’m just hyping you up. You’re the one who did all the work to get to where you’re sitting today.
So thank you. And I’d love if you could start by just sharing what initially drew you to the world of AI and chatbots.
[00:03:17] Kelly Mirabella: Well I’ve always been an early adopter of technology. I’ve been doing digital marketing since, you know, it’s been around. And I got into chatbots in 2017, really loved the idea of it became pretty recognizable internationally in the chatbot space.
And then I got a little burnt out after, you know, 2020 and things that happened around that time. So I took a step away from Doing chatbots for a while because I just felt like it wasn’t being as innovative as I like. I like to always be going after the next thing. And that was you know, a couple, couple years went by and I just did my digital marketing thing.
And then this thing called open AI released AI, like, As we know it today. I mean, there was a I obviously prior to this, but this is when it became a little bit more accessible to mainstream audiences and everyday people. And I was like, this thing is cool. I am just going all in. And I just wanted to learn everything.
You know, I had this insatiable hunger for To learn about AI and then a mentee of mine and a good friend, Ryan Baggott, who owns chatbot builder AI reached out to me and he’s like, I think that you would love to see how chatbots and AI can come together and I want to show you. And then he showed me and the rest is history.
I’m just like. Obsessed with this, so that’s how I got into it. I guess
[00:04:37] Mike Allton: that’s awesome. And I know, I mean, we’ve known each other for years. We’ve shared some stages. I’ve heard you talk. I’ve had you on some of our summits all about chatbots back in the day, to your point. So if you could kind of enlighten us, what’s the difference between those chatbots then and maybe some other engagement tools and the tools that we have available today, thanks to AI.[00:04:58] Kelly Mirabella: Yeah. When we’re comparing legacy platforms. To something like a, a Chatbot builder AI that really seamlessly integrates with tools like OpenAI’s ChatGPT, Gemini’s Google’s Gemini or Claude A nthropic. The difference is, is that number one, the chatbot is more intuitive to the human conversation. So instead of using traditional legacy style chatbot building, which is basically conditional logic. If this happens, then this happens. And these kind of chained conversations that you build out in the back end that become very big, really, what we can do is we can literally train an AI brain, if you will, we call it the knowledge base that acts like a prompt.
And a knowledge base that tells it, here’s what I want you to do. This is like your purpose. This is who you are. So you are the assistant for stellar media marketing. You are to do X, Y, and Z in this tone of voice respond concisely, no more than, you know, three, four sentences. And then below is all the information that you’re going to need to access about this company or what have you.
So I just put that all in the knowledge base, give a couple of examples, and then I’m going from multiple, multiple giant flows to one box. And the AI is handling it. And so when I go and I chat with, you know, I’m sure many of you out there have been the little chat box that will step up on a website and you start chatting and it’s very constrictive.
It doesn’t feel natural. It’s you know, press this button to do this, press this button to do this. And you’re constrained by how that chatbot is built. Whereas with the AI. I could say, Hey, I want to get a quote for such and such. And the AI can say, great, I need your name and your email. And I can say, why do you need my email?
And in traditional legacy bots, it’s going to break. It’s like, I’m sorry, that’s not your email. And you’re like, no, but I want to know why. And they’re like, I’m sorry, that’s not your email. Whereas with the AI, the AI is trained to say, yeah, you know, we understand, you know, privacy concerns. This is why we would request it.
And we want to be able to give you the information you need. Rest assured we don’t share your email. Are you ready to move forward? And this is all the AI doing this. And then the user me, I’m like, yeah, sure. Here’s my email. And then it’s still capturing that. So it’s, it creates a smarter experience and a more seamless and a better experience for our customers and end users.
And that’s just one of the many reasons AI is better. We can talk about how it’s. Instantly multilingual. It is able to do recognize anything. Basically you can go into chat GPT and use four Oh mini. We can do that in a chat bot and integrate it into our traditional ideals of how we use. Chatbots, whether it’s FAQs or lead generation, we can now do these things where it can change languages.
It can recognize images and give feedback if you want. It just comes down to how we as the builder train that chatbot. So it’s, it’s actually pretty amazing. I’m amazed every day what we can do, honestly.
[00:07:58] Mike Allton: That’s, that’s really, well, first of all, I was laughing at the example because it’s so true. It is true, right?
You’re like, hey,
[00:08:04] Kelly Mirabella: no, I don’t want to come at you. Oh,[00:08:06] Mike Allton: I hated using them. I hated building them. I know, you know, you were, you were fantastic at ManyChat and I was envious of people like you who could wrap their minds around these, these complex flowcharts and build them. It was too much for me. I don’t, I just don’t think that way.
Yeah, but. We’re first of all, we’re gonna have to have you come back another time just to do a whole show just about just to show you and really pull that apart. But I did have a question because I’ve used some of these web chat builders that are AI based and I really find them fascinating. The one I used, it was a very limited model language model and I was able to feed it like all the blog content that I’ve ever written just on AI and put on the website and people could ask it questions.
I haven’t created that much content about AI and marketing yet, so it still kept hitting situations where people were asking questions that didn’t know the answer to, but it was still really fascinating to watch. And the question I had about AI chatbot builder is how do we make sure that we’re not drawing on Information that is on the web and focused on information that’s in like a document brand, right?
Because people, they know about hallucinations, but they don’t, most people don’t understand where hallucinations come from. And it’s not so much that the AI is making things up. It’s that it can’t discern what’s true and what’s not. If it’s on a Reddit thread and somebody says something, it’s assuming it’s true.
And so it’s bringing in that kind of information. And how does chatbot builder get around that?
[00:09:34] Kelly Mirabella: First of all, I love this question. Two things that I have to say to this, first and foremost, we used to say with legacy builders, like your many chats and your mobile monkeys and chat fuels, right? These kind of old school, I said, you quote unquote old school.
Cause it was like 2017. Isn’t that old? But man, if 2017 is old, then you and I are in trouble. But we used to say that the bot is only, the chatbot is only as smart as the builder. And now we say that. The chatbot is only as smart as the prompter. So it really comes down to the prompt. We can absolutely create constraints and context around how we want the bot to access information, but to your point specifically, where if you have like PDFs that you want to upload and be like, I just want you to access this, it’s so wonderful that you just brought this up because I just did a demo.
I built it. So. Quickly, I am a assistant coach on two girls softball teams. One’s an 8U softball team, one’s a 10U, trust me, it all pertains. And one of the crazy things about being a coach, especially for two teams with two different rules, Like there’s so many rules. There are so many rules. We have a 30.
I think it’s like 35 page pdf of just like The rules and some of it is general it pertains to everyone some of it just pertains to au Which is one of the ones I coach some is just for tenure. So how do we like? In the middle of the game, someone’s like, how many feet between the pitcher mound and the home base?
Well, if it’s eight, you it’s 30 and if it’s a 10, you it’s 35. And I only know that cause I just trained my chat bot and it kept telling me the right answer. So I’m now trained too, but in the, in the clutch in the game, both the coaches and the refs have trouble kind of like what’s right, what’s wrong. And so I thought, Oh, You know, I’m a nerd.
I love chatbots. I’ve always wanted to test out what we call the assistant API. It’s an open AI. Offering that we have access to in chatbot builder. It’s already built in. You don’t have to like connect to anything. You go into your settings, you upload the PDF, you give it the prompt and how you want it to utilize the information.
And it’s now able to access the information you need. So I did too. I did one for 8U and I did one for 10U. We’re actually going to be rolling it out in our entire association. So all the coaches get to have this. So this is super fun for me. But basically I, I uploaded this 35 page document and I said, you are the coach’s assistant.
Your job is to answer questions in regards to, The eight you rules. I want you to access the answers in the document attached. There are general rules that pertain to everyone. And there are rules that only pertain to eight you you’re only focusing on the general and the eight you. And then I had to take it a step further because.
This is even crazier. Even though we have 35 pages of rules, not all the rules are listed because there is an, a body that oversees all softball leagues that also has rules. So then I had to create the contextual understanding in the chat bot that if you cannot find the rule in the primary source. Then you can access it by accessing the rules in the U.
S. S. S. A. or triple S. A. rules, which are just generally online. And so you have to, you know, tell it like when contextually important, when it pertains. You can access this other stuff, but I really want you to focus on, and we’ve tested it for both eight you and 10 you. And I was able to actually build the shop in less than 10 minutes because all you do is go in, give it the prompt.
I mean, obviously I know how to prompt, so it’s a little faster for me, but we also have tools that help you build the prompts. Like it’s a cheat cheat trick that we offer all our customers, but we put in the prompt, upload the doc and boom, I’ve got a chat bot. It’s already working. I can roll it out to everybody.
So You can absolutely tighten that control if that’s something you need. Whereas if I’m, let’s say I have another bot that I built as a demo, which is a coffee shop, like a, it’s a pretend it’s a demo. I just created this coffee shop. That is based in Durango, Colorado. Cause that’s where I went to college.
And I basically said to it, you’re this coffee shop is called Joe mama’s coffee. And you are. Going to answer questions about coffee. I don’t have to train the chatbot on all the different types of coffee. It’s already on the internet, right? I just have to give it the constraints of what does that mean?
If people are asking about coffee, the best types of coffee, coffee beans, if I’m selling specific products, I want to list that, but there you go. But then I also said to it. If people ask about Durango, Colorado, you can answer those questions. Well, initially that’s what I said. And so I’d be testing it and it would make every answer about Durango, Colorado, but it was like, you could say, what’s the best way to make, you know, what’s the best type of grind for a drip coffee or a French press coffee.
And the chatbot would say something along the lines of the best coffees for drip coffee in Durango, Colorado, like it’s Not part of the context of the question. So in order to fix that, I just went in and said, when it makes sense, when contextually people are talking about Durango, then you can add it.
Otherwise, just keep it coffee related. And the chatbot will understand it, but we have to just give that clear prompting. So like I said, just to recap this, Very long explanation. Your AI chatbot is only as smart as your prompt. So if you’re having trouble with the return when you’re testing, just fix the prompt.
There’s usually very small tweaks that we need to do that will fix it.
[00:15:07] Mike Allton: We spent a whole episode talking with Sunny Hunt from Convince and Convert all about the need to validate those responses, which was your point. If you’re not validating the responses, then you won’t know what needs to be tweaked.
But I’ve seen the same thing before myself, where I’ve fed chatGPT via Magai a bunch of these interviews that I’m doing for AI in Marketing: Unpacked. And I said, I want you to like identify themes because I want to start to build content based on the interviews. Like if we talk about advanced prompting and three or four interviews, I want to talk about that.
And I gave it some really fun guidelines, like identify the themes where at least three guests talked about these themes. And then we’re going to proceed to build some, some content around that. And I told her to pull quotes from. And when I looked, first of all, it pulled quotes from me and I’m like, okay, I had to tell it.
I am not the guest. I’m the host. I’m not going to quote myself in these, in these pieces. And then the second thing it did was it started to pull quotes, like from general internet, like it was quoting Chris Penn, but it wasn’t something he said in my interview. So I had to be very specific. And that was, you know, to your point, I put those guardrails in.
It has to come from these transcripts here, we’re not going to draw from the wider web. And as soon as I did that, it was flawless. So now I can create a whole library of newsletters and blog posts based on interviews and using ChatGPT perfectly.
[00:16:33] Kelly Mirabella: Absolutely. And, you know, I think that’s such a good example because People think AI is very smart.
It is very smart, but it’s also very stupid. Like you, the human are still a very important part of this puzzle. You can’t just throw things in like, like it’s going to magically understand the nuance of what you want, right? We have to really be clear. And I will give you a bonus tip. And this is something we tell all of our users when you’re building a chatbot.
Specifically a knowledge base is we, we have like the five P’s prompt engineering. And one of the last ones, I don’t remember what the actual P is, but basically it’s that we give examples. And by doing that, we will actually do so. I also like to say that AI is kind of like a toddler. If you tell it not to do something, it’s definitely going to do it.
So if I tell it, Don’t talk about politics. It’s 100 percent going to talk about politics. So if I want to say, don’t talk about things that don’t pertain, don’t talk about politics, don’t talk about religion, what have you, then I’m in my examples. I’m going to be like user colon. Who do you think is going to win the election?
And then assistant colon, I’m sorry, I am here to help you with X, Y, Z. We don’t discuss politics here. Let me know how I can help you with X, Y, Z. We give it examples. So anytime, whether you’re using chat GPT and you’re just trying to create, you know, these scenarios with your content, or you’re using a chatbot builder AI to build AI chatbots for your website or Instagram or Facebook, what have you.
You. You can almost solve every problem, honestly, with adding these examples. So if everything’s coming up wrong, you’re like, no, no, this is the example. This is how I want you to answer. And it will, it will definitely pick up on it then.
[00:18:12] Mike Allton: That is a great tip. Let’s let’s go high level for a second. How are you seeing like the entire like customer service and marketing interactions?
How are you seeing that being changed by these kinds of AI chatbots?
[00:18:26] Kelly Mirabella: Okay, specifically to what I do, a lot of people see that I’m like the chatbot lady, right? But my business, Stellar Media Marketing, is a digital marketing business. So my clients are people that I am doing content marketing for, social media.
I’m scheduling stuff out using Agorapulse, which by the way, I was in Agorapulse all morning scheduling like a month’s worth of my client’s content out. Yeah, I love it. That’s just how I roll. But once once it’s in, what we can do is connect a automated system to like this, these AI chatbots to, let’s say Facebook as an example, and we can create number one call to actions right on Facebook.
The post, we could be like, you know, say the words pool down below. I have a pool client. Say the words pull down below and we’ll get you a free quote. And so it creates this much easier, seamless experience. So there’s that, but that’s something we could do before. But what about the AI that makes it better?
Two things. Number one, my AI, if you go to stellar media marketing, my Facebook page I think it’s stellar 24 seven. Or to my Instagram, my Instagram is way more fun to be honest. And you comment on anything, my AI assistant will respond and it will respond individually to what you are saying in the tech context of my business and with the tone of voice that I’ve trained it on and what this does at a really high level, nerdy, Algorithmic understanding, especially if we’re talking about Facebook and Instagram is it sends a signal to these platforms that says, this is an engaged post, which means people are interested, which means we should be showing it to more people.
And it actually grows your reach. And I have done numerous testing and studies on this with my clients. I could show, not right now. I can’t show numbers because I don’t have. You know, in a file, but if I wanted to, I could pull up numbers. I could show screenshots. I’ve done it before that show that when I take over someone’s account, their engagement, their reach, their like every metric you look at as a social media manager and the client that is hired by them is going skyrocketing because my AI chatbot is trained.
To engage, whereas you know, social media manager is not going to catch everything or they’re a little slower to get to it because they’re human and that they should be slower. Take the weekend off social media manager, my goodness, right? And that’s why AI chatbots can help. But on the flip side of that, the other thing we can do is if someone does come into the chat from these places or from an ad.
I have a video coming up where I’m going to teach people how to do messenger ads that connect to your chatbot. The chatbot, as I mentioned before, is much more intuitive. I can train it to capture specific types of information when the context is showing that a person wants a quote or whatever it is they want.
I Maybe it’s a lead magnet and they ask about it in the chat. The AI has been trained to recognize, Oh, this person wants this lead magnet about whatever, I don’t know, using AI chatbots. And I’ve trained it to ask, what is your name? What is your email? Here is the PDF. And it will collect that. And again, if it gets interrupted, cause they’re like, well, why do you need my email?
Why do you need my phone number? Whatever. The AI chatbot’s still going to take care of it. So you have more chances of capturing those clients and especially if they come in and they’re speaking Spanish or Chinese or Russian or whatever language, there’s like 50 plus languages supported by the AI. And it will flip to it right away.
And it still recognizes this person wants this thing, a quote, a PDF, whatever they want. And your job is to get this information and then give them the thing that they want. And it will absolutely do it. And then it will notify you and your client. If you set it up that way, Hey, you got a new lead. Let me tell you, my clients are always very happy when they get that little ping, you got a new lead.
That’s, that’s helping my business thrive too.
[00:22:18] Mike Allton: That’s awesome. I just want to reiterate. What a difference that is from the tools that we had just a few years ago, where if you were doing automated comment replies, you sucked because your bot sucked because it was nice. Nice comment. Thank you. Yeah,[00:22:34] Kelly Mirabella: robotic.
Now it’s custom. Like I have one person would be like I had the one that comes to mind was a, someone that was like a dog page. Like the whole page was dedicated to their dog, right? On Instagram. And they commented, but it was like in their tone of voice with like a little rough and my AI chapel, it’s like, yeah, rough, rough.
You’re so this, yes. You know, like they made it into that vibe, but still answered the question. For my business. So it just makes it more fun. And if you go and look at my comments, you’ll know, like my AI chatbots on it. I also really quick. I love this example just pulled in yesterday. My I’ve also trained my AI So like I had someone messaged me on Instagram trying to like sell me fake followers.
I don’t even check these messages, honestly, but my Did you see the story? My chatbot’s like, it appears that you’re trying to do something that is against the rules, but if you would like to actually learn how to build your following the right way, feel free to reach out. Like it was trolling the troll.
It was so funny. I was like, yes, yes. High five, high five to me.
[00:23:40] Mike Allton: Right, right. That’s awesome. I need that for my LinkedIn. Inbox.[00:23:46] Kelly Mirabella: Oh, man. I wish we had LinkedIn chatbots. They haven’t released the API for that yet. Unfortunately, I don’t even check my inbox on LinkedIn, bro. That’s scary.[00:23:57] Mike Allton: Yeah, but I love that your bot is not just, you know, replying in your voice, but they’re matching and mirroring.[00:24:03] Kelly Mirabella: Yes. It’s very NLP.[00:24:05] Mike Allton: I love that. That’s fantastic. Now you, you, you’ve been talking a little bit about your customer success stories. I’d love if you could share a specific example. You don’t have to get into the details in terms of data. That’s okay. You don’t have that. But could you share a little more about a specific example of[00:24:20] Kelly Mirabella: how this has[00:24:21] Mike Allton: really improved their presence?[00:24:23] Kelly Mirabella: Yes. I have a client right now who sells pools. This is the backyard pools. One thing that we sat down and talked about before I signed them as a client is leads. And I think that all clients are saying this, and this is actually from a perspective of a social media manager. This is a hard conversation to have because honestly, as someone who has mentored many social media managers and been doing this a long time, we can’t guarantee leads.
But it’s becoming more and more important in order to keep a long term client that you can prove your worth. And one of the ways that I know I can do that now, before I couldn’t, now I can because of my AI chatbot. So, pools, it’s like 100, 000 for a pool, right? Anyway, I think their lowest price pool is maybe 600, 000.
65, 000, right? So this is not a small T shirt project kind of thing. This is a big ticket. And so not only have we increased their engagement, we being me, me and my AI chat bot, we not only have I been able to increase that, those numbers of reach and, and like, when I tell you, I increase it, we’re talking like.
A thousand percent. We’re not talking 20%. We’re talking insane numbers because my AI chatbots doing the heavy lifting. But on top of that, we started running ads and just low budget ads. I actually have a video on my baby got bought YouTube channel. That’s coming out. I think the 24th of September I’m like booked out, you know.
Plan ahead. I’m a batcher. I love matching. But I actually show, so if you really want to nerd out on this, like go check out baby got bought subscribe. Cause at the end of this month, there’s a video to show you this exact thing I’m about to go through. So I’m doing a a traffic ad that leads to an, instead of a website, it’s leading to messenger and I’m targeting it 25 a day.
On the weekend, the one we’re setting up in the video is for Labor Day weekend. So basically a four day weekend or three day weekend, it runs today, it should be running right now or in about an hour, and then it will go until Tuesday morning. And I’m finding that between usually it’s just a small weekend.
This one’s a longer weekend. It’ll cost us a little bit more, but 25 a day. You’re looking at between 50 and 75 for the weekend. This is a very low budget ad. Usually you would never see leads. Like ask any ad manager. They’re like, yeah, that’s impossible. Maybe one, if we’re lucky, we’re getting six to seven leads every weekend that we run these ads.
And I know for a fact they are leads. Why? Because they are coming in through messenger, my AI chatbot, I can see it all happening in my inbox. My AI chatbot is taking over, is answering all their questions. It’s qualifying the lead. It is the person’s like. I want to get a quote and my chatbot is like, what kind of project are you looking to do?
And it’s taking in that information because they do pools, they do pergolas, they do outdoor kitchens. Great. Give me an idea of the size of your backyard. What’s, by the way, what is your first and last name? What’s your email? Can we get the address of where the project is so that we can add you to our database and give you a call?
And set up your quote and we’re getting all this information. And if there’s ever that feedback of like, why do you need this? AI is taking care of that. And then we’re like, great, we’ll be reaching out to you on Monday because we know that this is a weekend ad. And then it’s sent immediately. It’s sent to three places.
Number one, it sent a notification to my client. Direct to, they have it to email, but I can also do it to their phone. Number two, it’s coming to me because I don’t want it to accidentally slip through the cracks. I’m like texting my client. Oh, Hey, did you see this lead that just came through? Cause that makes his world happy.
He’s just, yeah, thanks. We’re on it. I’m just kind of laying down that, Hey, did you see what I did for you, sir? I’m out on, on the lake this weekend. And you know, when I get that ping, I’m like, do you see what I got for you on my holiday weekend? And then it’s also going to Google sheets. You can have it go anywhere you want.
Like it could go to your CRM, you can connect that, whatever, but this is the system that we have set up. So. I can do that. Like that’s done, but I’m still not the one who has to be on doing the work. We can enjoy our weekends. We can have our cake and eat it too. We can enjoy our weekends and still be able to provide these services to our clients.
And they’re honestly, I have two. Clients that I have done these things with and they’re just blown away. They’re thrilled with what we’re able to do.
[00:28:50] Mike Allton: Okay. Fun fact about me. I used to sell swimming pools and hot tubs and billiard tables.[00:28:55] Kelly Mirabella: It[00:28:56] Mike Allton: is. Well, you know what? It was one of the funnest industries I’ve ever worked in because standing in that warehouse, we, we, we bought a Michael’s and, and it was a whole chain of stores back in Ohio, Pennsylvania.
People came in. Looking for fun. They didn’t came in cause they have to, right? No, no, no. You don’t buy a pool. Oh, you
[00:29:17] Kelly Mirabella: buy[00:29:17] Mike Allton: a pool cause you want that. And it was actually really great sales training. Cause we would, we would spend a day a week doing sales training. So we understood how to, how to frame it, how to talk about the outdoor experience and not focus on the width of the pool wall.
Cause nobody cares. So that’s fascinating. And I, I, I would have loved to have had this kind of a system bringing me leads on the weekends. Yeah, make your
[00:29:41] Kelly Mirabella: job easier.[00:29:44] Mike Allton: Guys, you gotta pay attention. We’re talking with Kelly Mirabella about how brands and marketers can use these kinds of AI chatbots, not only manage, but even improve the customer engagement.
I want to talk about how we get started with chatbots and maybe even address some of the ethical considerations. But before we get to that, I want to share with you the tool I’m using every single day as a thought partner to talk through strategies and ideas. This episode of AI in Marketing: Unpacked is brought to you by Magai, your gateway to making generative AI incredibly simple.
Wondering how to seamlessly integrate AI into your marketing strategy without getting bogged down by complexities. That’s exactly where Magai shines. It provides user friendly AI solutions that empower marketers just like you to innovate and elevate your campaigns without needing a degree in science.
Imagine having the power to generate creative content, insightful marketing data analysis, or even personalized customer communications all at the touch of a button. Magai isn’t just about providing tools. It’s about transforming your approach to marketing with AI that’s tailor made to be straightforward and effective.
So whether you’re looking to boost your content creation process or want deeper insights into your marketing performance, Magai makes it all possible with a few clicks, no fuss, no hassle, just results. Ready to simplify your AI journey? Visit Magai today to learn how their solutions can revolutionize the way you engage with your audience.
with your audience. Don’t just market, market smarter with Magai. Tap the link in the show notes. So picking, backing off of that example, we were talking about, you know, of relatively low. I mean, you’re right. That was like almost no advertising budget at all.
[00:31:23] Kelly Mirabella: It’s nothing. We’re spending like 500 a month.
Like that is, Yeah, or this process that I put together with AI chatbots, it would never, it would never provide anything like that.
[00:31:35] Mike Allton: Yeah. So, but for businesses who think, oh yeah, I could afford that, but maybe are concerned about the AI chatbot itself. If you’ve got a smaller budget, there’s thinking, oh gosh, this is going to cost me 10, 000 a month to do this kind of AI chatbot What would they be looking at?
How do they get started with a smaller budget?
[00:31:54] Kelly Mirabella: So with chatbot builder AI, which is the company that I work for. And honestly, like, I know I may be biased, but I could tell you, cause I’ve used pretty much all the chatbot tools out there. It is the best when it comes to AI chatbots, it’s seamless. And you don’t have to pay for all the extra bells and whistles, like.
Open AI. You don’t have to pay for your own API to use that in your bot, which you have to use because that’s how you make your bot smart. It comes with it. Right now we’re priced at 49 a month per account and that’s unlimited contact. So no matter how many people come into your bot, unlike like a company like many chat will charge you more based on how many contacts.
I mean, people, we don’t do that. And that includes the integration with open AI for many everybody gets that it includes the world’s best customer service. We have the best customer service team. So honestly, it’s, it’ll blow you away, but I will. Do a little note here that we actually are about to raise our prices.
So I don’t know when this podcast goes out. I can’t disclose how much we’re raising our prices, but we are raising our prices soon, but if you get in before we raise our prices, which will be the end of September your grandfathered in for life with the 49 a month. So it’s an, even with the price.
Increase you still, we’ve calculated, we’ve done like a side by side comparison to other brands. When you start. You know, if you really want to make a smart chatbot, you have to get your own API for open AI that costs so much. If you want to integrate other tools that come natively in our account, if you, you know, compare that all of our accounts come with unlimited context, con, not context contacts that we don’t charge for that.
It ends up actually being significantly, if not 50 percent less than our, our competitors. Even more in some cases, so I think it’s worth it, obviously. But it doesn’t cost that much if you think how much it helps you. And by the way, also our platform uniquely is not only on all the meta platforms, like you could put us on Facebook, Instagram, WhatsApp.
We also do telegram. We have a native website integration that fully customizable to the look and feel of your website that also has an instant. Voice, so you can actually talk to it and that is all built in as well. So we, we support so many more businesses. We actually support Google, my business as well, but Google, my business just changed their messaging.
So we’re kind of on hold till they transfer over to their new system. But we cover a lot more bases.
[00:34:29] Mike Allton: Yeah, I mean, I think even if you were putting like 100 bucks a month into this tool and[00:34:35] Kelly Mirabella: 900[00:34:36] Mike Allton: bucks a month to advertising a grand total across your social and your web, this is a huge and powerful and a very affordable investment that is going to reap all kinds of rewards because you can be using this both for sales and for customer support and retention and, you know, giving your customers that easy Instant access to not just a static knowledge base, right?
But that conversational back and forth that’ll understand their questions, answer them, make them happy. Your retention can also transfer
[00:35:06] Kelly Mirabella: over to a human very smoothly. We could do a chat history. This is one of our favorite features. We have chat history. So if you want to, you can actually have the AI.
So. Summarize the conversation and give sediment scoring and a summary not only a summary, but also recommendations on how to improve that conversation. And that is whether you’re chatting or you’re using the call, like we have income, we could set up an incoming call chat bot with like voice and everything, which I just did videos to show how that works below your mind.
It’s so cool. But that same thing we can. Basically, you have the A. I. Grade summarize and give recommendations on how to improve, which you just don’t see anywhere else.
[00:35:48] Mike Allton: So we’re using these A. I. Chatbots for for outreach, for engagement on social media and our website for customer support. Maybe we’re actually using it for sentiment analysis and summaries as well.
What are some of the ethical considerations that marketers need to consider when they’re using chatbots like this?
[00:36:04] Kelly Mirabella: I think anytime that we’re using AI, we need to be very clear that it is AI. So if you were to chat with my chat bot, it introduces itself. My name is Stella, the AI assistant for stellar media marketing.
I think a lot of times business will go, especially new, new builders will go into this and they’ll say, Oh, I don’t want to tell them it’s an AI. You know, I, they won’t like customers won’t like it. I’m going to tell you right now, customers don’t care as long as they get their answers, like. All the data points to this.
They would rather the chatbot just fix the problem than have to wait and talk to a human. Cause most times the human is less helpful anyway. So we need to be very straightforward with setting that expectation that this is an AI chatbot and not only that, but legally in many places. And I believe in the near future, all places, it’s going to be a requirement.
So you might as well just get into the habit of saying, Hey, this is the AI assistant for Speller media marketing, Kelly Noble, Mirabella, whatever it is, how may I assist you today? And I think that that’s super important. The other thing that whenever you’re building chatbots you have to make it easy for them to opt out.
Most platforms like a meta already have that built in where if anyone hits you know, types the word stop, it will stop. We have SMS built into our system. We use Twilio and If you’ve ever received a text from someone that out of the blue you should not have received, just type stop. If they don’t stop, that is against the law.
So like you have to have the systems in place to make it number one, obvious that this is not a human they’re talking to. And number two, it also needs to be easy to opt out of. The other thing I would say, you know, to the those people who are like, well, if it’s AI, they’re not going to want to talk to it.
It’s going to chase people away. In my experience with a well built AI chatbot, people forget very quickly that they’re chatting with an AI chatbot. We’ve actually had customers reach out to us and we have like reviews that are saying, this blew my mind that this was AI. Like I am so surprised I actually forgot.
And by the way, if you ever want to test one out and just like have fun, go to chatbot builder AI and chat with that chatbot. And there’s an Easter egg, ask for Snoop Dogg. And your mind will promise you, you can actually talk to Snoop Dogg. It’s wild.
[00:38:22] Mike Allton: I love this. And, and, but to your earlier point about being transparent, I’m doing the exact same thing with my LinkedIn newsletters.
So I use AI to generate a weekly newsletter that summarizes that week’s podcast episode. I’m very transparent, this is written by CLU, my AI assistant, and people love it. It’s got 2000 subscribers on LinkedIn, knowing that it’s, it’s
[00:38:43] Kelly Mirabella: People don’t care as long as they know and it’s useful.[00:38:48] Mike Allton: Exactly. And that’s the thing I got to make sure with this newsletter, that it’s, you know, providing an accurate summary, it’s drawing real quotes, and that’s exactly what some people want to see.
They’ll glance at that, they’ll take 60 seconds to read it and be like, Oh, okay. That is or isn’t an interview that I want to listen to. That’s great. So in the few minutes we have left, I want you to take out your crystal ball and see where we’re going with this stuff. And it’s funny because we’ve talked a lot on previous shows about, you know, AI and understanding context.
We kind of touched on that a little bit today. AI helps me draft the questions for these shows. And it often says like, how do you see AI changing in the next five to 10 years? Like, that’s the question I have to ask you right now. And I’m thinking, okay, that’s ridiculous. How about five to 10 days? Can we go to 10 days?
How do you see chat bots changing?
[00:39:38] Kelly Mirabella: Really quick story. I went to social media marketing world years ago. This is like 2017, 2018. And they had, I want to say Molly Pittman and. A couple of people on a panel talking about chatbots at the time, nobody was talking about AI that did not exist in our world.
And the, the question person, the host of the panel asked the question, where do you see chatbots in 10 years? And the entire audience was laughing their butts off. So there you go. But in the next 10 days, I could definitely talk to that. It is moving at the speed of light. Like the future, there’s a book called the future is faster than you think, because it’s.
It’s so fast. And the way technology stacks is that as technology gets smarter and humans use technology to get smarter themselves. And now we have AI, artificial intelligence, kind of fueling the fire of the speed of technology. If you think about how technology has moved through history, we’re talking about decades.
It’s measured in decades, years, months. Days, weeks, days, minutes, seconds. It’s already changed while we’re sitting here. So it’s so hard to predict what’s next because it’s going too fast. Like literally we’re sitting here and I guarantee you, my CEO is implementing something new in chatbot builder because.
Something just became available to us. With that said, I do think that evergreen wise, like, you know, what is the grand scheme of what it could look like if we were to fast forward ourselves in a year, even I think it will be drastically different than what we can even conceptualize today, what’s possible.
I do think it will be worked into just about everything that we do online, kind of the internet of things already. If you go in and you use. Google’s like suite of products. Like I’m a huge Google suite of products person, my Gmail calendar, my docs. And I am a beta tester for Gemini’s assistant, which essentially Gemini is the open AI chat bot, chat GPT of, you know, Google.
And I’m able to now currently use it. Let’s say I’m in a spreadsheet and I don’t know how to do a formula, and I’m like, Hey, can you add this formula? It will do it for me. So it becomes an essential part of the experience in that suite, and I see that happening more and more. You look at, Microsoft is already utilizing co-pilot and a lot of their products, so it will be.
A must have, or a already, it’s not even a must have, it just is, right? So anyone who is skeptical about AI today, and because AI is going to take our jobs. Well, I don’t argue that AI is not going to take jobs. I think it’s definitely going to take the jobs of people who refuse to adopt the technology and to try at least to understand the concepts of what it can be used for.
Because it’s, it is inevitable. That is how I see AI. It is inevitable. I do think in the future, we’ll see more government stepping in worldwide and creating constraints around it, which I know that a lot of the AI companies don’t want. They claim because innovation will be stalled, but I do think that it can be very dangerous and not because AI in and of itself is dangerous, but because, and I don’t want to freak people out at the end of your show, but AI is more than just.
These chatbots that I build or the, the OpenAI’s GPT chatbot or Magai, which by the way is one of my favorite AI tools on the planet. Just a little to them You know, to create content or analyze things. Yes. Wonderful. AI. It’s also used in medical practices right now in studies where they’re able to quicken the process of testing and coming up with cures like they’re actually already talking about cures for cancer because AI is helping them or diagnosing cancer patients long before like breast cancer patients long before.
They even realized that it could, like, nothing’s even developed, but the AI is helping to recognize these things. These are life saving things. But on the flip side of that, AI is already being implemented today in governments military, defense systems, and it is, Kind of terrifying to think like they’re like the US, for instance, has a like a governing body to look over the military militarized AI use cases, and they have a rule that says, like, You can’t attack a human, you can’t kill a target without human intervention, but not everyone has that rule.
So I do think that there’s going to be bigger impact of AI than even people are conceptualizing today. It’s, we’re already seeing it in the politics of America and like the content that’s putting out the deep fakes. I think that as a society we need to recognize it’s very powerful. It’s inevitable, but we also have to keep our critical thinking, you know, so there’s a lot, it’s a lot to take in.
I do think it’s going to be an integral part of every part of society, every part of life. And I think people need to adapt quickly in order to stay in a place where they can. At least stay safe from being tricked or falling behind or whatever that looks like depending on the AI that you want to look into.
That was kind of deep and scary. I’m sorry.
[00:44:59] Mike Allton: No, but you are a hundred percent right. That’s the whole premise of this show, at least for marketers. And I’ve talked about on my blog, I’m going to do a whole episode about how we are in the fourth industrial revolution.[00:45:10] Kelly Mirabella: A[00:45:10] Mike Allton: lot of people don’t understand that, but you know, there have been, you know, Points in our human history where our entire society, every aspect of society has changed so dramatically.
And this is the fourth time. It’s going to be one of them,
[00:45:23] Kelly Mirabella: right? You know, like the tractor cam, everyone’s like, Oh, that’s the end of farms. That’s not the end of the farm, the farmer, that’s the end of the way we did it before. Right. So does it end everything? No, it’s just, we adopt as we always have throughout history because of technology.
The wheel didn’t change the fact that we. Travel, it just made it easier to travel,
[00:45:47] Mike Allton: you know, right, right. Yes. Some people who, you know, who made their livelihood taking care of horses and stables, you know, those people had to adjust and transition, but our society as a whole, most people would argue is far better as of the invention of, you know, the automobile and so many other things that came around with the first industrial revolution.
So absolutely. Thanks for bringing that. And thank you so much for all the knowledge and energy you brought today. I know folks are gonna love listening to this and they’re probably gonna want to go back and listen to it a couple more times because you’ve been amazing for folks who have more questions and they want to reach out, where can they go?
[00:46:19] Kelly Mirabella: I am everywhere at Stuller 24 seven. I am on. The Ticky Talks, Facebook, Instagram, YouTube. If you wanna learn more about chatbots outside of you, go to stellar media marketing.com or Stellar 24 seven.com. So everywhere, stellar 24 7. Or if you wanna learn specifically about Chatbot Builder, who I’m the COO for, please go to chatbot builder.ai and we have free accounts now that you can play with.
So. Go check it out.
[00:46:48] Mike Allton: Fantastic. We will have all those links in the show notes. And if folks, if this is your first time kind of thinking about AI, don’t forget, I have an AI primer for marketers on the social media hat. com. I’ll put that link in the notes too. It’s going to walk you through everything we’ve talked about today and more so that you understand chat GPT and open AI and large language models, and these kinds of concepts that you don’t necessarily have to be an expert at, but you should be familiar with as a marketer so that you understand what’s happening in the, In this industry with this revolution, and you can come out ahead until next time, welcome to the grid.
Thanks for joining us on AI in Marketing: Unpacked. I hope today’s episode has inspired you and given you actionable insights to integrate AI into your marketing strategies. You enjoyed the show. Please subscribe on your favorite podcast platform and consider leaving a review. We’d love to hear your thoughts and answer any questions you might have.
Don’t forget to join us next time as we continue to simplify AI and help you make a real impact in your marketing efforts until then, keep innovating and see just how far AI can take your marketing. Thank you for listening and have a fantastic day.
Related
Discover more from The Social Media Hat
Subscribe to get the latest posts sent to your email.
