They Didn't Ghost You. You Were Too Slow.

What's the real cost of a slow lead response?

Leslie Tracey: Hey everyone, back with another amazing guest on the Diamond Hands podcast. We have Shannon Blake. She's the Director of Aesthetic Partnerships at Podium, and she's led that partnerships role for the past six and a half years at Podium, which is an operating system for aesthetic practices. So today we're going to talk about this: most aesthetic practice owners know they're losing leads — I'm sure you've lost leads before, I have too — they just don't know exactly where, when, or how much it's costing them.

As I mentioned, Shannon has spent six and a half years leading the company's charge into the aesthetics vertical, and they launched the first AI operating system purposely built for aesthetic practices. We're not going to be doing a product demo here — if you want a demo, you'll have to call them yourselves. What we're going to talk about is how you can evaluate what questions to ask and what outcomes are actually realistic when you're implementing AI into your practice. So welcome, Shannon.

Why do practices juggle five disconnected technology systems?

Shannon Blake: Thank you so much for having me, Leslie. I'm excited to jam out with you for the next little while.

Leslie: Let's talk about the operational reality that a lot of our listeners are going through, because we have beginners listening and we have more senior people listening who are running practices. The average practice runs three or more disconnected systems and wastes eight hours a week switching between them. Can you walk us through what a typical day looks like for a patient care team and where their time goes, and how the owner can't even see any of that on an analytics report?

Shannon: Yeah, absolutely. The average aesthetic practice does have — we actually see closer to at least five technology platforms that they're using. They typically have their EMR, which is where they do their scheduling, payments, and charting forms. Then they've got their CRM, hopefully, to manage all their leads. Then they've got their phone system, and maybe even a different texting platform for text marketing. In some cases they may have a platform that's got more intense inventory management tracking. It's a lot.

We've been heavily focused on aesthetics for the last three-plus years, and we started realizing, with the advancements in AI, that there really isn't a reason to keep letting aesthetic practices deal with this fragmented technology hell. When you've got a CRM that's not talking to your EMR, your leads are coming in and going into an email inbox that's not tied to your EMR, it gets messy. And how it all plays out is you're missing leads and losing revenue, to simplify it. Luckily, thanks to the advances that changed the game, it's now easier to build technology, which means we're moving into a world where we don't need fragmented tech anymore — it's going to be easier for companies to build one all-in-one solution. Thankfully we got to that point, because it's about to make these practice owners' lives much easier.

Leslie: I've seen it myself going into practices — you have five screens open, you're looking up rewards on this one, rewards on that one, and it's a lot. I purposely worked the front desk at one of my practices to see how much they go through in a day, and I was like, this is a hard job.

Shannon: It is. And in many cases, you don't have those screens with you on the go. The reality with an aesthetic practice is there are always leads coming in and things happening, and not having it in your pocket on a phone, and the usability of running the business on the go, that's also another challenge.

Who actually owns following up with a lead?

Leslie: And from an accountability standpoint, who is responsible for tracking all of these things? Who is responsible for updating them? It creates an accountability mess too, from what I've seen.

Shannon: It does, a hundred percent. Here's the reality: leads are coming in, and the second you see one come in, you ask your team, "Hey, did you remember to call them, have you followed up with them?" And they're like, "Oh yeah, we'll get to it." But you don't really have time to follow up, or to actually know if they ever got followed up on. That's why I'm excited to talk about what AI employees can do today to help solve that, because there's no scalable way for a human to manage and track leads. There's the first touch, and then there's the ongoing follow-up that gets missed even more than the first lead, because a lead doesn't come in and instantly respond to you, and you respond to them, and they pick a time to meet and get on the schedule all within seconds. That's not how it works. It's always a long, drawn-out process, and leads and conversations go cold.

Leslie: And that's a big deal. I know there was one practice I worked with that had an Excel sheet. They'd get a lead by email from their vendor, and they had this spreadsheet where they'd mark down if someone had been called the first time, then a second and third call went out. By the time you call this person the second time, they've already gone somewhere else where it's easier to book, easier to get on the schedule, and you've just got hundreds of leads sitting there. It's crazy.

What does secret-shopping aesthetic practices reveal about response times?

Shannon: Leslie, I've actually lived this. We attend quite a few conferences throughout aesthetics, and one of our favorites is Terri Ross's summit. Because it's a more intimate event, we've been able to secret-shop every practice that's attending — we text them during business hours as well as after business hours. If they don't have a textable number, that's a big no, that's a whole other conversation. But if they don't, we'd fill out the contact-us form instead. I firsthand was texting these practices during business hours and after hours, and it was scary what I found. Average response time during business hours is actually getting pretty good, about an hour and a half on average, which is okay. After hours, it's eight and a half hours.

Leslie: And we all know why that is — no one's processing those after hours.

Shannon: Exactly, we're trying to have a work-life balance here. But eight and a half hours is far too long. And the ones that don't even have a textable number go to an email inbox, where I wouldn't hear back for days. Any lead flowing into an email inbox that you have to sift through to find the actual leads among the junk — you already know you've got a broken process. You're not going to be able to meet consumer expectations, which is two minutes or less, period. Every lead should get a response in two minutes or less.

Leslie: It's interesting that you say that, because with this there's relational trust and there's efficiency trust.

Why has the threshold for "fast enough" collapsed?

Leslie: The threshold for efficiency trust has gone down so much. Ten years ago, we'd be okay if we emailed a place and they took 48 hours to get back to us, because we were moving on a slower timeline. For those of us who are older, geriatric millennials, we remember booking a hair appointment and thinking, okay, she might get back to me by the end of the day tomorrow. But now, with AI, people's brains have been trained for instantaneous replies, and if they're not getting one, at least for booking, they're going somewhere else. So, a lead comes in at 9 p.m. on a Friday after the practice closes — what happens to that lead, and what's the actual revenue cost of an eight-hour response time versus a two-minute one?

Shannon: I'll answer that, but I want to say, your patients have been programmed to want convenience even before AI — Amazon, Uber, Uber Eats, DoorDash, Netflix, all of those companies changed the game. I blame it mostly on Amazon, honestly, but it's the world we live in, and it's the world your patients live in. AI didn't really come to market in a way that could be truly valuable until 2022 at best, so the good news is AI has advanced enough now to help solve that problem.

What happens when a lead comes in after nine? I know this because I've done the secret shopping — it goes unanswered until your practice manager or front desk shows up the next day and has time to log into all the systems and find wherever the leads are coming in from, which is about eight hours or so, depending on timing. And if you think about the loss in conversion rate already, it's a 40% increase in conversion on leads if you can respond in two minutes or less.

What makes a booked appointment actually stick?

Shannon: You've got your own average ticket price for your services and treatments, and you know the lifetime value of your clients, so do the math yourself. If you could get a 40% increase in lead conversion to booked appointment, what does that mean for your practice from a revenue perspective?

Leslie: It's meaningful, it's really meaningful. Let's bring it back to the real world, because it's one thing to talk about it in theory. What will increase the stickiness of someone actually keeping that appointment? Some people are in different camps about charging for a consultation or not, because what also happens when people are so quick to book is they're also quick to ghost. No matter how fast they book, these people with immediacy might look somewhere else, find a better deal, and book the next day with someone else and forget to cancel with you. So how do you make that process more sticky in practice so they follow through to the booking?

Shannon: I'd love your thoughts on this too, but what I'd say, coming from a company that's built AI employees, is that the experience the lead has from the minute they're excited about something matters. That conversation they have with your AI employee at nine o'clock at night — we call ours Avery, but whatever AI employee you're using — needs to be a wow experience. It's not just about being fast at getting back to them, it's about how informational your AI employee can be. Is it getting them excited about the treatment? Is it asking questions about how long they've been thinking about the procedure? Is it building excitement about the outcomes they'll get? All of that happens in the magic of the conversation, and then you're creating a very convenient experience, because your AI employee is saying, "How does tomorrow at three o'clock sound?" "Yes, that works," or "No, what about Thursday?" Having that fast back and forth with no delays matters — when you've got their attention on scheduling, you want to capitalize on it right away.

It gives them, without even realizing it, a kind of dopamine hit: this was a really easy interaction, I got all my questions answered. That person — who doesn't always know it's an AI employee, which is the goal — was so helpful and so excited for me. Credentialing the providers, if that's something you do, really pushes trust in who they're going to be meeting with. And of course, appointment reminders are critical, and getting the forms out, which again AI can handle automatically so humans don't have to. Those are the little points along the way that increase the stickiness of the appointment. If they're filling out forms and confirming their appointment, and they also had a really good experience on the phone learning about the treatment, all of that plays into increasing the likelihood they hold the consultation.

Does Avery handle phone calls or just texting?

Leslie: So Avery — you said "the phone." Was Avery talking about that, on that text conversation?

Shannon: Avery will be talking — we actually rolled out Avery for AI voice, and it wasn't quite as dialed in as we wanted, Leslie, so we rolled it back and we're working on her. But right now she has an incredible messaging voice, she's very informational.

Leslie: That's great, and I love that, because Gen Z has a lower average ticket right now, but hold on, ladies — in the next five to seven years they're going to pop. This also goes back to service mix with Avery, because the people who are going to be texting a lot are going to be Gen Z, saying "I want to get in for a facial" or "I want to get my lips done," and they're going to be talking in that type of way. Whereas I might still call — I'm about 40 years old — I'd call and see what I'm getting, and I'd wait until business hours because I work during business hours. I'm not up at 10 o'clock; I'm sleeping when some of these Gen Zs are booking appointments. So I think you have to differentiate who's more likely to use text message to try to get an appointment versus who's going to call, and the data shows it's our younger patients, at this current time.

How can AI improve phone training and team reviews?

Shannon: Since we're talking about all things AI, there are some really cool features where you can program it to your perfect phone conversation. If you're an owner, you've probably given some kind of phone training to your front desk, because that's such an important touch point — you need to surprise, delight, and be helpful and excited.

So you can train on the back end: here's everything I want asked on the call, here's what I want them to say, bring up, not say, information to gather, and the tone I want them to use. You train the AI on the back end, and then every phone conversation is recorded, transcribed, AI-summarized, and scored. As an owner, getting a report on the effectiveness of phone conversations by individual gives you quick insight into where the gaps are or where a call is falling short, and makes sure that conversation represents your brand and the way you want your clients spoken to. That's another cool thing, because no practice owner today has time to listen to all the calls and then go train after that. But if you could just have a report served up to you — here's a concern call you might want to listen to, or here's a great call to share with the team as a best practice — it makes the practice owner's life so much easier, and they feel like they have a better pulse on how those conversations are going.

Leslie: I agree with that, because everything stems from your vision, mission, and values. If you don't have those dialed in first, everyone in your practice, including you, is going to be shooting in the dark. If you do have them dialed in, you'll know how you want people to be talked to, you'll understand what an ideal phone call is like, and you'll understand what type of clientele you're trying to attract. That's also going to change how you speak to them — if you want all Gen Z, you're speaking to them differently than if you want high-net-worth people in their 40s and 50s who might be spending $20,000 or $50,000 a year in your practice. I think it's really important that you brought up reviewing phone calls, because that's one of the things many business owners in our space have trouble with — carving out the intentional time, whether it's the practice manager or the owner, to say, "I'm going to review how my team is performing," and bring it up at a weekly or monthly meeting to talk about wins and areas of improvement.

Shannon: I want to layer on, because it's fun to share where AI is good enough now and where it isn't. Another cool feature, since you just brought up one-on-ones and team meetings — with a true all-in-one solution, AI can make your one-on-one PDFs for you and pull in whatever you want to highlight: utilization rate, re-booking rate, how many Google reviews came in for someone, literally anything. It gets whipped up and prepares you so you look like you have it together, ready to highlight their wins and talk about areas of opportunity. I'm guessing in a lot of cases those one-on-ones aren't happening because it takes so much time to pull the information together and prepare. That's another cool thing AI can do today.

Where does the machine stop and the human take over?

Leslie: That's a big thing. Where does the machine stop and the human have to take over, because it's one thing to be able to book easily and all of that?

Shannon: It's a fine line and a balance we're still figuring out, but from our experience, AI before 2022 wasn't good enough to manage a human conversation lead. That's all changed. If you have a good AI employee, built off the backs of some of the most respected large language models — like an OpenAI or an Anthropic model — it can sound more professional, more on point, more accurate, more on brand, even than your own employees, because you've trained it to do that. I'd say 90% of messaging can be handled by your AI employee. Where a human needs to step in is if someone is asking for medical advice — you don't want your AI employee getting into that. Or, if you have a good enough AI employee, there's a way for her to raise her hand: Avery, our AI employee, will raise her hand if the conversation has reached a place she doesn't feel comfortable continuing, or doesn't have the information to answer intelligently, and a human steps in. That's where humans come to play — the consultation, welcoming them into your practice. I don't see a day where AI replaces those human interactions and connections; that's what makes this industry so special, the relationship building, clients feeling like they're part of your family. That's something I don't foresee us ever wanting AI to take over. But anything task-related — scheduling appointments, rescheduling, answering the same old questions about treatments, sending out forms — AI is good enough today to handle, if you have a good AI employee and you train it.

Leslie: I'd 100% agree with that, because people do want ease these days. I also think it goes back to your client mix — some people will be upset if they know they're talking about their cosmetic needs to an AI, and they'll go somewhere else. How do you deal with that? Some practices, unfortunately, don't train their AI properly, and it gets very templated, and the person of means gets turned off because they're a bit more old school. And then there are some Gen Z who are actively saying they want a slower life — I've seen 16- and 17-year-olds with flip phones, because they feel too hyper-connected to all of this. Can you talk about that from an ethical and humanist standpoint, because there are some people who are just put off by this too?

Shannon: They are. There's a lot of controversy around what AI is doing for the environment and how much power it takes, and I get it — I commend these younger kids, their hearts are in the right place, and I understand why they're fighting against it a bit. I don't have a perfect answer, but I'd tell you, you need a system where, if you know you have clients who want that human touch and don't want to talk to an AI, it's easy to reassign the conversation to the person you want owning it, and the AI won't touch it. That's fully customizable — if you wanted anyone under a certain age engaged with a human, or above a certain age, that can be set up. But honestly, we've had an AI employee out in the market for almost two and a half years now, and it's pretty rare that we get this concern, which is surprising to me. I do think it'll be a growing, changing thing. But to this point, Avery gets Google reviews about how amazing her customer service was. People walk into practices asking for the AI employee, and staff have to tell them that wasn't a real human — we hear more of that than frustration. Back in the day, when we called a business and knew it was AI, we'd say "representative, representative, customer service" just to get out of it. With AI employees, at least in this industry, if somebody ever says they want to talk to a human, it's an easy hand raise from the AI employee into the software, and a human jumps into the conversation. But it's honestly been quite rare for that to be requested, because the technology is that good.

Leslie: It's also a market-fit thing — the people seeking these cutting-edge technologies are usually cutting-edge people themselves, and their brands are trying to be more cutting-edge too, so they're attracting patients who are more okay with that.

Shannon: Totally.

Leslie: So I think it's all a bit of a pendulum. It just depends.

What's behind the 45% lead conversion case study?

Leslie: There was an OpenAI case study of Podium's AI agent that showed a 45% increase in lead conversion and a 30% increase in revenue for practices that adopted it. Based on your perspective, what specifically is driving these numbers, and what did practices do during implementation that created that result? Because it's not just "we bought this" and it's perfect on the other end — let's talk about the dark work.

Shannon: Yes. That goes into setting these things up, because I've seen it across all sorts of systems — it is dark work. Leslie's correct. I'll work backwards on this one. Here's the thing: AI employees do not come out of the box ready to slay it for you, I'll be honest. There's some understanding, obviously — there's a lot of backend training on the aesthetics industry as a whole, so the AI employee knows the lingo, acronyms, services, and treatments — but you have to take the time to train your AI employee as if it were your own employee.

Luckily — we call ours AI Studio — there's a very robust section where you can upload all of your processes, procedures, and playbooks. If you have a brand playbook, everything. You can feed it into her brain. But after you've trained it, you want to play with it and make sure there isn't anything you've missed. So you have to put in the time, once she's trained, to have your staff texting in and pretending they're a lead to try to break her. You'll find little things where you're like, I didn't think to train her on that. Try to break her and give her bizarre questions. At Podium we do have onboarding managers whose job is to get all that information and help you get it in, so you're not fully on your own, but I do try to prepare people that there is post-testing training you need to do.

And then you need a high-level understanding of how to self-train her. We make it very easy, luckily, so it's not difficult, but you have to carve out the time to spot-check her every once in a while and then know how to go and train her. I don't want to paint a picture that it's so easy, that it's just going to take five minutes and she's going to be ready to convert all your leads. It takes training. But the technology is at a place where, if you're willing to put in the time, your AI employee can be pretty rad and pretty on point with your brand.

Why does "garbage in, garbage out" apply to training an AI employee?

Leslie: And I think that's great, because we have to be honest about what AI can do and what it cannot. At the end of the day, it's the same thing I learned in finance, recruiting for financial planning back in the day when I used to do that: it's garbage in, garbage out. So you have to take the time to collect all those different pieces, those data points, and input them into the system. Going back to mission, vision, and values — if you don't truly understand what your mission, vision, and values are, if you don't truly understand what your brand is from an operational perspective and a voice perspective... I always say, treat your brand like a person. How does that person show up? Are they classy? Are they ratchet? Are they young and sassy? Your brand should have a voice. Everyone you're hiring, if you're hiring to vision, mission, and values, should embody that too, in their personalities and the way they operate. And if you know that so well, then you put the proper things into the AI. But I think the challenge is that many practices have not done that work. So all they can say is what is wrong, but they can't say with complete conviction what is right.

Shannon: Yeah, I think what's helpful is, if you go out and you're shopping for an AI employee, you want to ask: where can I self-train it, and how do I make adjustments? Because you all know you're constantly evolving your business. You're either adding a new service or treatment, or maybe you're adding a new membership program. You have to start thinking of this as a real employee in your practice. Would you bring on a new service or treatment and not go train your employee that you have this treatment and how to speak about it? If you were adding a new membership, you would never not train your front desk on that. Everything you're training your humans on, you need to take the time to also go in and quickly train your AI employee on, so it's getting the same information your humans are.

What drives the revenue lift from an AI employee?

Shannon: And Leslie, to go back — I didn't answer what you asked me about that OpenAI case study. The increase in revenue is very meaningful, and I would point to a couple of different things. One, we've talked about a lot, so I won't beat a dead horse, but it's how fast you're getting back to your leads, so you can lock them in while they are most excited about the treatment. Two, having an AI employee follow up with leads that go cold, because sometimes they get to a point in the conversation where they're like, I need to check my funds, or I need to talk to my partner, or whatever. There are so many distractions in our lives. Having an AI employee take that on: okay, I've been trained to follow up on a conversation if it goes cold after 24 hours or after 48 hours, however you want to customize it.

And then, if you have an all-in-one solution with an AI employee, you can ask things like, "Send me a list of individuals who have shown interest in BBL Moxi but never booked an appointment," or "Pull me a list of individuals who have shown interest in X treatment, or only got this treatment once," when we know the best outcome is if they do it three times every year, or whatever it is. It's being able to revenue-mine in a very fast and effective way, and then work with your front-facing AI employee like Avery to send out the messages and start the conversations, to get one step closer to getting them back on the books. So it's using AI to revenue-mine, because you have a lot of lost revenue just sitting in your database today that no human has time to comb through, and AI can do it in seconds.

When is an AI operating system overkill for a small practice?

Leslie: And that's true. So let's talk about a solo practice owner or a small two-location practice. When is it overkill for that size of business? When is the right time to implement something like this?

Shannon: I would say if you are a brand-new practice with very low lead volume, nothing overwhelming that you can't handle white-glove by yourself, you don't need a full-blown system like this yet. I think the focus needs to be creating your brand, getting your beautiful website up and running, and working on generating leads, whether that's through hiring a digital marketing agency or whatnot. It's when you get to the point where you're no longer able to get back to all your leads in under two minutes, and you no longer have time to follow up with leads that have gone cold, so you're losing revenue that way. That's the point where I would recommend looking at a more robust system that has AI throughout. But if you're a small single owner-operator just getting started, we always tell people that until you've got about two revenue-generating providers, your volume probably isn't so out of control that you can't manage it yourself. For a larger practice, it's a no-brainer. There's probably so much lost revenue that has just slipped through the cracks, because you haven't been able to follow up when you're so high volume. That's definitely where it makes sense to have something like this.

How can a newer practice use general AI tools like Claude?

Leslie: And from your perspective as an AI thought leader, can you talk about how a smaller practice can use AI — not this system, but whether it's Claude or Claude Cowork or whatever — to make their lives easier or templatize things? Tell us more about that process, from a patient coming in, to getting the treatment, to sending the treatment follow-up plan, when someone's building and they have fewer than 300 patients.

Shannon: Oh my gosh, I would say go and sign up for Claude literally today. It's not that expensive. It could write a business plan for you if you wanted it to, but you can also brainstorm with it on different marketing promotions you can run as a new practice. You could have it write the descriptions for your website on different services and treatments. It can make you a website — my friend just spun up a website in a few hours. It's incredible what it can do today. There's really nothing it can't do. It could literally be your best friend and business partner that you can ask: "Hey, I'm starting a med spa. Here's what I do know, here's what I'm passionate about. Help me write a mission or a vision for my practice." And it can be very interactive. It's not like you're just farming it off to AI. It's a launching pad for creativity, where you can jam out and go back and forth until you land on something amazing that feels good for you. It doesn't have to be just a command and an output. It's, "Okay, I like where your head's at, but let's tweak it here or there." I use it every day, even in my own job. It writes my bios for podcasts.

Leslie: Love it. I like to say that in this era, we're almost more like editors now, because if you're taking anything AI gives you and just using it — I'm sorry, I'm going to say it to you — you're lazy.

Shannon: Yeah, 100%.

Leslie: So you're using it and editing it based on your eye and your judgment, and you have to do that. Even one use case: if you're a newer practice owner and you don't have the funds yet for bigger things, you could feed AI all your treatments and say, "Make a treatment plan template." Then it's, okay, this person is this age, whatever, and based on what you put into it — again, garbage in, garbage out — it will spit out: this person needs to come back four times for this, and so on, and create a whole treatment plan for you. Do not put any patient information in there, though. There are so many cool things you can do if you don't have the bigger systems yet that can make your life easier. It could send you your schedule for the week automatically. Even though I have an assistant, I also have mine send me my schedule for the week, and it says, "Leslie, you have this, look out for that, prepare for that." I'm like, okay, thanks.

Shannon: Yes, girl, we could go on and on. You could say, "I want to run an event — give me some ideas on a good treatment package to highlight there." Or, "I want to be more involved in my community and have better referral partnerships with other like-minded businesses that might have a similar customer base. Can you put together a list of businesses I could partner with that would be mutually beneficial? Draft me an email I could send over to them." Anything. It's incredible, the creative assistant you have in your pocket with AI now.

Leslie: Exactly. And again, I can't harp on it enough: make sure you edit it. Because if this owner gets an AI email from you and they know it, they'll delete it.

Shannon: We know, we know. I get LinkedIn connections and messages and emails — girl, I delete them immediately when I know it's AI, a copy and paste. I just know it. There's something in the syntax, how it's written.

Leslie: And sometimes they even paste it where you can see the little highlighter behind the text, and you're like, that, literally.

Shannon: I know, I know. It's crazy.

How long does implementation take, and what should you expect after 90 days?

Leslie: But let me ask you this. What should a practice realistically expect 90 days after implementing an AI operating system correctly? How much staff time does it really take? And what should they expect as far as leads converted and patient follow-up, compared to where they started? Give me the true implementation timeline.

Shannon: Yeah, so the implementation timeline is going to be 30 to 45 days, as long as your team is available and making it a priority. That's come down — some migrations in the past have taken a couple of months, so it is coming down. But I'm not going to lie, it's still migrating systems, and it's a pain in the ass. The reality is AI is not going anywhere, and what you can do with it to make you more money is a hundred percent worth going through migration hell. So I'm just going to say that. You're still switching everything over from one thing to another, so it's still going to be a painful process. We try to make it as seamless as possible, and you have hands-on support throughout the whole thing, but plan on 30 to 45 days, multiple trainings and meetings, and working with an onboarding manager to migrate information over.

And then you asked me what to expect. After 90 days, your AI employee and your full migration are rocking and rolling at this point. You're going to see about a 30 to 40% increase in lead conversion as well as revenue. But again, that's only if you're using the platform the way it's supposed to be used. When I tell you AI can pull you a list of leads that showed interest in a BBL Moxi treatment and never booked, you've got to go and prompt it and ask it for that information. Or: "I want to grow my memberships by 20% next month. Send me a list of individuals who are a no-brainer to sign up for our membership program based on how often they're coming in, and then draft a text that you'll have Avery send out to re-engage them and get them to sign up for the membership." You have to engage with your AI assistant, ask it things, and have it do things for it to revenue-mine, if that makes sense. But Avery, the front-facing AI employee, is the one responding to all of the leads, and that's from day one. As soon as she's trained, she's converting leads for you, which is increasing revenue, obviously.

Who needs to own the implementation inside the practice?

Leslie: Yeah, but I think it depends on the size of the practice, and again, it depends on capacity — I like to be very realistic. It depends on capacity, because if you have a busy, busy, busy practice and you really don't have any employees with the capacity to clear their schedule — and that's what I'm trying to be very clear on with everyone—

Shannon: Yes, thanks, Leslie.

Leslie: —the capacity to really clear their schedule and say, "I am going to be the implementation commander for the next 30 to 90 days."

Shannon: I'm so glad you brought that up. There does need to be someone who is a point of contact and is available for things like, "Hey, we just migrated over all of the appointments. Can you go in and spot-check a few for us to make sure we've captured everything?" It's software, so any software is going to have its little glitches here and there. There has to be someone who can surface those, then follow up and make sure they're resolved. I'm so glad you brought that up, because if you don't have someone who can dedicate, I would say, four hours a week for the first month—

Leslie: Girl, I'm sorry, I've got to cut you off there. I have seen this happen even with just an EMR — it is not four hours a week. I'm sorry.

Shannon: That's what it's been for us, but with other AI operating systems, maybe not. It just depends on the migration — what can be migrated with AI and what can't.

Leslie: Because I'm like, four hours a day, honey. It's a full-time job for the person who's actually doing the things. I've seen people working around the clock with Podium to get this stuff done. So I'm just letting you know, because I try to be very realistic — I ain't trying to lie to anybody out here, and you're not either — but I've seen Podium be implemented in a practice, and it took someone's life away for about 30 days.

Shannon: The timing of when that happened matters too, I will say that, because if this was before May, very different.

Leslie: A hundred percent, it was before that.

Shannon: Yeah. We have been on a mission to figure out how to make it a smoother process, where there's less the practice owner has to do, because originally, migrating from an EMR over was a whole new ball game for us. We have been humbled and we have learned a lot. Some systems are easier than others to migrate, but honestly, now that we use AI to pull everything over, I feel like it's much better. When you're training Avery, that's more of a nuanced scenario. If you've got everything already documented and you can just upload it, that's easier, but if you've never documented anything around your different processes or playbooks, then it takes a little longer, because you've got to create those.

Leslie: Exactly. And I think that's the key to be realistic about: okay, I'm happy implementation takes less time, but you have to have that foundation — the playbooks, the SOPs, the language, the products, how often they should be used, and all those different things. Because Avery needs to know if your protocol is that you want someone to come back every three months for this or every six months for that. If you don't have that written down somewhere — these are all our services and these are all the use cases for them — that's going to take time.

Shannon: Yeah, you've got to have it all documented and figured out. But it will feel so good once you've put pen to paper and figured it out, and then you can easily upload it into an employee that's going to be consistent, accurate, and on brand for you.

What three questions should you ask an AI vendor before signing?

Leslie: Yeah, and I agree with that. So there are steps to this, guys. The last thing I want to ask you: a practice owner is sitting across from a technology vendor now, because you guys aren't the only ones offering AI to aesthetic practices right now.

Shannon: Nope.

Leslie: What are the three questions they should ask before signing anything?

Shannon: Oh man, I have to limit it to three? I literally have a slide of about 15 questions to ask.

Leslie: Give us the first three, then send me the resource and I'll drop it in the podcast notes.

Shannon: Amazing, amazing. Okay — and we've learned this the hard way. Ask if you can self-train the AI or if you have to go through their team to make edits. When we first launched an AI employee, you had to go to your client success manager to make any edits, and that is a nightmare. I'm telling you right now, you'll find yourself adding and tweaking so much that you need to be able to do it yourself, and it needs to be easy. So I would ask, "Can I self-train it? Show me where and how. I want to see how easy it is to make adjustments." That's one.

I would also ask how long they've been building AI employees and AI agents, because we've been at it for three years, and you guys, it is hard as hell and it is very, very expensive. We've been very lucky to raise $450 million, so we have the money, but I'm telling you, as someone who works very closely with our product team, it's not something you just whip up real quick and it's amazing. It takes reinventing it over and over again. Avery's on year three, and she's pretty damn good, but if they've only been building an AI employee this year, I'd be very leery, because they have a lot to learn. So ask them how long they've been building AI employees, to make sure you've got someone reputable.

And then a third — I would say HIPAA compliance. That maybe needs to be the first. Have them talk you through whether the AI is HIPAA compliant, because in this industry you can't mess around with that. I know there are two sides, where some folks believe that if it's cash-pay aesthetics, it doesn't need to be HIPAA, that it's more for insurance. But we always play it safe: aesthetics is healthcare, and everything needs to be HIPAA compliant. So make sure their full platform is HIPAA compliant, but especially their AI employee.

What foundation should a practice have before investing in AI?

Leslie: I think that's great. Oh my goodness. So guys, I'm going to have everything about Shannon Blake and, of course, her company in the show notes, so if you want to reach out and get more information, do it. And as a base, remember: if you want to do this, make sure your vision, mission, and values are really clear and that you have your SOPs laid out. Understand all your products and services and how often you usually recommend them to people. If you have that base, and you have a good number of employees, it's time to look into a better solution if you can't support the leads that are coming in. So Shannon, thank you so much. This was fun.

Shannon: Of course, thank you so much for having me, Leslie. Your energy is so contagious, so it's been fun hanging out with you for a bit.

Leslie: Yes, and I'll see you soon, I'm sure. Until next time, guys — if you liked this episode, make sure you leave a review.

Shannon: Thanks, y'all.

Leslie: Want more clarity like this? Join the branded practice newsletter community. I write about strategy, systems, and brand leadership, straight to your inbox. Go to diamondhandsmedia.com to sign up. And if this episode hit, share it with someone still building without a brand. This is the Diamond Hands podcast. Thanks for listening — now go lead.