Transcription
Boom shakalaka. Welcome back to the Austin Automates YouTube channel. In this video, I'm going to show you how I turned this solar database reactivation campaign into something new and improved. This one wasn't using any AI, and I actually am using Closebot version two and recreated the whole thing using AI and the AI logic.
Now, again, this is the old version. This doesn't have any AI. This is actually what I was selling to solar companies when they wanted to turn their older cold data into new appointments. I had this down to a science. I knew that if I ran 1,000 leads through this, I could book anywhere between 10, 15, 20 appointments depending on the quality of the lead list that they gave me.
Now, the problem was there were a few issues that I had inside of this. And those of you trying to screenshot this and, oh my gosh, I'm going to recreate this—you could do that. I do have a place that you can go and actually get this entire solar snapshot. I give away this workflow, and I also give away my reputation management workflow. I used to upsell. I used to tell people, "Hey, I could get you 10 appointments, and I'll throw in a month free of reputation management." I give away both of these inside of my community called GHL Agency Blueprint. Right inside of the classroom, I have trainings very specific to database reactivation, reputation management, and inside of the database reactivation one. If you go to solar snapshot, you can actually get the solar snapshot. I have one for personal agency, dentist, and all and concrete and all that stuff. But what I want to do first is show you how I turned this into this and made it much, much smarter.
So, there were a couple things that I kept running into when I was doing database reactivation for solar companies that I always wished—man, I wish this function worked a little bit better. I wish that AI could actually help out. Now, a lot of you think—it's probably think—it's in the conversational pieces of it. It's not. At the end of the day, having static messages, I sent out one open-ended message. It would wait for people to reply, and then it would send them down these paths. Yes, positive intent, which positive intent very rarely worked. I had someone tell me to burn in hell, and it sent him into positive intent, and I was like, "Really? Depending on who you are, that might be positive; that might be a little bit negative for me—not so positive." And then if they said no, it would send him down a "I totally get it," and then it would retry with a message, and then if they said yes, no. Now, the conversational piece wasn't where I wanted AI to step in; where I wanted AI to step in was this conditional—were the conditions. So, as I sent thousands and thousands of messages through this workflow, I would build on to it based on actions that I had to take manually. And people would respond, and then whenever they would respond and they wouldn't fit into one of the conditions, I would have to add the conditions in there. So if someone said yes, sure, yep, positive intent—like I said, it didn't really work really well. Then I would get no, nope, not. And I started getting this thumbs-down emoji quite often. And so I built that into the conditions. Not right now, not at this time, not now. Wrong. Already I have with Nova, which was the solar company I was working with—have solar already installed, stop, and then contact reply false. And what would happen is a lot of time—most of the times—this worked; however, there were times that people would say things that it didn't match with one of the conditions I set up, or someone would say yes but follow up with me at this time, or, "Hey, I'm interested, but I have some questions," and I really wished that that I could use AI conditionally to then separate people because it would be a lot smarter, and I didn't have to like always be monitoring whether people went in down the right path.
And so this is how this worked. It sent out a very open-ended question: "Hey, we spoke. Are you still open to getting solar?" This framework works for every industry. "Hey, we spoke at one time. Are you still interested in blank service?" Wait for a reply. Then when they reply, they can only say so many things: yes, no, burn in hell, go away, oh, super interested, I'm super happy, blah, blah, blah, blah, blah. And then obviously those that don't reply, they would then get sent down another path where they would try again. And so this right here was my money maker for a while. And then I would tell people, "Hey, listen. I'll book your first 10 appointments," absolutely, without you paying me a dime. Once I book your 10 appointments, you—you build my—you do my buildout fee. And then the cool thing is I charge $2,000 usually to set this up. But I also have an AI bot that can turn your older cold customers into five-star reviews. What I want to do is I want to set that up for 30 days absolutely free. Then I would set up this workflow, my reviews workflow. And bada bing, bada boom, I would send their previous customers in the review workflow, and then I would send all of their leads that never converted into this workflow. And what it would do is then separate all of their leads into different pipelines and all of this. But there were things that were broken. The conditioning—the conditioning was always where this thing failed.
And so what I did was once I found out that Closebot version two—look at the difference—the different. So, there are a couple things that I still want to add to version two, but I'm—I'm going to dive in—dive a little bit into what's exactly happening here. So, you saw in the conditions of my solar one that I had to write in like the phrase—is containing—the phrase contains a certain word, and every once in a while someone would slip through. With this, it's very simple. So, it starts with the open-ended question: "Are you still interested in getting solar put on your home?" And then I have the—and then I have the switch op option or node—or I don't know what these are called. I'm sorry, Bryce, if it's a node, then it's a switch node. I'm going to say switch option, but I have the switch option where I can use AI to power the decision description. Are they interested in getting solar put on their home? And then yes, they're interested; no, they're not interested; we have the wrong number; already have solar. So, the cool thing was is once it already has that, if I publish this and I test it, I'll trigger the bot, and it says, "Hey, guest"—obviously that would say first name—"I hope you're having a great day. Are you still interested in getting solar installed on your home? Let me know if I can assist you." And I say, "Yeah, sure." Now, AI logic moves them down that path, and I don't have to go and add sure to the conditions, which this just makes it so much easier. Great. "Would it be okay if I ask you a few questions about your home to see if you qualify for the solar program?" Fine. Because that is also conditional where now it's going to start the actual questions pre-qualifying them and then take them through certain objectives. Awesome. "Does your house get full sun?" And then it will take them through the objectives. So, what I have is the switch. The switch says yes, they're interested. And then it takes them up here where it says, "Can I ask you a few questions of your home, about your home?" If they say yes or anything along—yes, now is a good time or right now is not a good time—then yes, it then takes them through the objectives. And then once they get the objectives, then it actually books a call with them to talk to a solar representative. If they say no, now is not a good time, then it will book a time for them—a better time to follow up. Now if it's—if they say, "Hey, are you still interested in getting solar put on your home?" and the switch is no, they are not interested, then I can—again, this—the logic is where the magic happens here because it would—this is where it would always break inside of my solar one because I'm like—I had to like constantly be guessing, are they interested or not interested? Is that what that means? And now AI can like very intelligently say, "Hey, that's yes; that's what they mean—not interested," and then it would hit them with my second question: "Hey, listen. Totally get it. A lot of people, they weren't interested because they didn't save money. Are you—if we could save you money, are you interested?" Then a lot of people say, "Yeah, sure." Then—then it takes them through the objectives of pre-qualifying them and then booking the appointment. Then if they say, "No, they're not interested," "Totally get it. Would it be okay if we followed up in the future?" This was golden because this is where a lot of magic happens in database reactivation. If they've said, "Hey, can I get solar on your home?" No. Okay. "Hey, you want to see the new program?" No. Okay. "Can we follow up in the future?" Sure. So, people don't like saying no three times. And so that's—I found that out while sending thousands and thousands of text messages through this one. Like, people would always—like if I asked them, "Hey, could I follow up in 3 to 6 months?" That'd be fine. Now, I don't have that built in quite yet. And again, there's still some things that I'm going to be adding to this, but this is like the barebone structure of how I've done database reactivation. And every single database reactivation campaign I run looks very similar to this. I've done this in solar, roofing, HVAC, real estate, now a cement company, fitness facilities, direct sales professionals. I know there's one more; I always forget one, but whatever. But I always use the exact same framework: open-ended question, and then the logic then decides based on their responses which branch they should go down. And then at the end of it, are we turning off the conversation? So people that say wrong number, it says, "Hey, sorry about that," and then it just stops responding. And then what I'll do is probably add something on here to tag them inside of HighLevel or delete their contact because we don't really need that. And then moving people through. But at the end of the day, these switch nodes—I really hope I'm saying that—these switch nodes or switch options where they use AI as logic, that's where the magic happens because I don't have to recreate this and build this all out. But that's—that is a brief overview and breakdown of these—of how this works. Now, stop responding is obviously pretty self-explanatory; just the thing stops—stops responding.
Now, objectives work very similar to the objectives in—inside of version one. I like to keep them linear just because I like very specific questions being asked at a specific time. You can stack objectives by adding different—adding objectives and then just stack them. That's completely up to you. I chose not to do that; I chose to do it linearly. And then conversation. So, like this one right here: "Hey, did you—" if someone says, "I already have solar," and, "Oh, excellent. Did you have a great experience?" Because what I like to do is actually try and pull a referral out. If they didn't have a good experience, then they'll have a conversation with the bot about what went wrong, what was their experience. Branching—again, I'm still learning this, but branching seems to be like more of a—is this true or is this false? If you look—use power of AI—the description, it literally gives you a true or false option, a left value, a right value, and so yeah, that's more of—it's like the switch, but you get two options instead of one, and then you just highlight and hit delete if you don't want to use that. And then there's some delays, custom web hooks, verify address, get property details, GHL booking, and then custom scenarios. I'm not going to go into custom scenarios too much right here, but right inside of this same workflow, I can build out a custom scenario that says, "Customer says they want to talk to a human," and talk to a human, and now I can build out a custom scenario that if any time during here they say, "Hey, I want to talk to a human," this will fire the custom scenario, and I don't have to go and build that out inside of a completely separate workflow, and I can just tell them, "Hey, right on, we're going to get you—" Whoops, I can tell them, "Oh, it reset. I can tell them, 'Hey, right on. We're going to get you connected,' and then set up like a web hook to—to then trigger some sort of message. Again, still working out some of—some of how all this is going to work.
And then the next steps—adding to this—is what I want to do is just like I had in my old one is every single pathway updated—um—a pipeline step. So it put people in a pipeline of yes, no, not right now, wrong number, have solar, stop, didn't respond, and it would move people through pipelines. So the next step is to add that into this flow where no matter which direction it's going down, it's actually updating a pipeline because this was real—this was really important for the solar companies that had call teams to be able to see. And I'm going to continue to make this smarter. Now, the cool thing is I do have a call with Bryce on May 5th where I'm going to build this out to the best of my ability and how I would want it to work. And there's a couple things that I want it to do that I'm not sure if it can do yet. But Bryce and I are going to spend 3 hours on the phone, and he is going to help make me—help me turbocharge this. We are going to slap a turbocharger onto this database reactivation bot, and I'm going to make it a little bit more generic so that people can plug and play this entire bot into whatever industry they want. And whether it's roofing, HVAC, I like targeting businesses that have door-knocking teams because they always have an influx of new leads. That's a little hack of mine; I don't know why I'm telling you that. But me and Bryce, 3 hours, we are going to turbocharge this thing. And he's going to help me get it to do things that no one else in Closebot version 2 is going to have access to. And all of the people that are inside of the GHL agency blueprint are going to get exclusive—you are the only ones that are going to get access to that call. And I'm going to give you—I'm going to give you this bot. And I'm also going to do a very specific training on solar inside of this group once this is done. I already have a few solar companies lined up that I'm going to be testing this with and running this through. I worked out—I do not ever recommend doing a rev-share deal, but I do have friends that are still—that run solar companies. And I was like, "Hey, let me test your AI out. Like you pay for text messages, you pay for the software, but I'll do all the setup, and then you just pay me on the back end on deals that close." I don't recommend you ever do that unless they are like a tight-knit friend that wouldn't want to lose you as a friend if they didn't pay you. The cool thing is now I could use AI to follow up with appointments and stuff like that and find out if they did close, which was hard to do that in this.
So, if you would like access to this when I do release it, and you would like access to this right now, um, and you would like access to the call that I have with Bryce on May 5th, jump into GHL Agency Blueprint. I do have the link in the description below. And if you sign up for HighLevel or Closebot through my link, you can see that in the database. Let's actually go to reputation management. You can see that my affiliate link for HighLevel is inside of these trainings. It's also on the about page. If you sign up for HighLevel through my affiliate link or Closebot through my affiliate link, I will hop on a one-on-one call with you and help you get set up. I also do five live support calls inside of this group, and currently we are right smack dab in the middle of me updating my DBR training, and we did the sprint training day one, depending on when you're watching this. All of this is inside of GHL agency blueprint, and that's how you're going to get access to this. Please reach out if you have any questions. That is how you—that is how I'm building out a database reactivation, and that's a teaser into version—Closebot version two and how to set up a DBR campaign in there. And again, I'm not making this up; I've been doing this for years. So, if you would like to have database reactivation training and the templates and workflows and snapshots from someone that's actually done this, like I said, jump inside the group, and you get all of that. You guys have a great rest of your day, and I look forward to seeing you in future trainings. Woo!