Brent: All right, and welcome back here, day three at Cloudways Prepathon 2024. AI and Innovation is our theme for today. I hope those notebooks you all have at home are just packed full of insights, gold nuggets, takeaways, things that you’re going to implement for this year’s Black Friday Cyber Monday and just general holiday sales traffic season that we’ve got coming up here. Some awesome suggestions there on automation from our panelists.

So next up we’re going to be talking about unlocking some growth, leveraging AI in e-commerce for online store success. And so our speaker, Colin Lepiscopo, he’s the co-founder of AI Advantage, an agency. He is an SEO pirate, an AI Navigator, a power blogger, a Brazilian jiu-jitsu blue belt. He loves helping brands, building friendships, and training in self-defense. He also is a disruptive innovator, a Maverick, and he’s spent at least 12 years of experience helping for-profit, nonprofit brands grow their business by showing them their blind spots. Colin, I’m pumped for this presentation, I know our audience is too, so let’s give Colin a nice warm welcome here at Cloudways Prepathon 2024.

Colin Lepiscopo: Awesome, yeah, I’m super stoked to be here. Thank you guys for inviting me, and I’m joining virtually from Sunny San Diego. It’s 11:00 a.m. here, so people are just waking up to go surfing, so that’s kind of where it is right now. And I love just virtual events because you don’t have to wear pants. Who’s not wearing pants right now, anyone?

Brent: I can’t answer that in my official capacity here, Colin, with Cloudways and DigitalOcean, but maybe we can let our audience answer that.

Colin Lepiscopo: Yeah, I’ll let them answer. You guys are weird. Okay, I’m wearing shorts. All right, I’m just, no, it’s okay, I’m just kidding. I’m not kidding about the shorts, but it’s okay if you’re not, I get it, you know, we’ve been doing that for a while now.

So why should you care about who I am? You probably don’t know who I am, but thank you for the awesome intro, Brent. I’ve been doing marketing for about 12 years. I was director of marketing for Social Media Examiner, which is a big conference actually here in San Diego at the convention center. Really passionate about MMA and self-defense training, so trained a lot of Brazilian jiu-jitsu. Anybody else train or do MMA here? If you’re not familiar with jiu-jitsu, that’s Ryan Hall doing an Imanari roll and then catching, um, I think that was BJ Penn, he caught BJ Penn in a backside 50/50. And anyway, so rolling around on the ground, the portion of UFC that’s boring, that’s what I do for a hobby. Beats the treadmill.

So 92% of companies have said that they are using AI driven personalization to drive growth. Okay, so not even just AI tools or AI solutions specifically, personalization. So type yes in the chat if you are currently using any type of AI solution, and no if you’re not. And if you said no, I’d like to know why your ex boss did jiu-jitsu. Okay, can’t talk too much about jiu-jitsu and MMA, otherwise that’s what I’ll do the entire time. So okay, some of, I’m seeing a lot of yeses, so that is good.

So let’s just talk through what we’re going to cover today, what’s in scope. Okay, I want to explore basically how AI is going to address the unique challenges that you have coming up for this holiday season. So there’s key areas where you can unlock some growth using AI tools, and then I’m going to show you some real world success stories. So it’s going to be some practical applications for e-commerce, and I’ll give you some recommended tools and experts to follow, and then I’ve got some free resources for you at the end. All right, so if you’re already bored, I’m sorry, just hang out till the end, you get some free resources. It’ll be worth it, I promise.

All right, so let’s just briefly talk about the current landscape as we’re prepping for this holiday season and some of the unique challenges that you have, and the pressing issues and the opportunities that you have with those challenges. So we already know high competition, right? You’ve got stores vying for all of the attention of customers. So standing out can be tough. Your PPC ads get more expensive, and it’s just really really difficult to get your content in front of your customers. There’s also managing expectations, because there’s expectations for faster shipping times, better deals, bigger discounts, exceptional customer support. These are higher than ever and they’re just going to continue to get better as certain brands are raising the bar. And then there’s still many stores are still facing difficulties managing stock levels, handling those supply chain disruptions.

So you’ve got your traditional approaches to these problems, and then you have some AI driven approaches. So here’s how you can address them traditionally: it’s manual customer service, generic marketing campaigns to just quickly blast things out everywhere, and then a lot of reactive inventory management. We know what this is going to lead to, right? It’s all those problems, it really doesn’t help. The AI driven approaches, you can use some custom AI chatbots and some personal sales assistants, you can use some personalized marketing and Predictive Analytics.

So here’s some of the benefits of actually doing this. You’re going to enhance the customer experience, so AI provides personalized experiences that are going to cater to individual needs, and this results in higher satisfaction and loyalty, and what we see in some of these stats which I’ll share here in a little bit. And then we were just talking about automation, right? So you want to improve efficiency, automation of repetitive tasks, right, that’s what this whole last session was about. It’s going to free up your team to focus on more strategic initiatives. And then you’re going to use AI driven insights that are going to help you make better informed decisions, whether it’s for marketing, inventory, and customer service. I’m all about data driven decision-making, I’m a big CRO guy, so I’m all about experimenting and using data to inform good decision-making.

Okay, now let’s look at some practical ways to do this, and how you can actually unlock some growth for your store during this holiday season that’s coming up. Personalization can lift revenue by as much as 15%, and 69% of customers said they were more likely to purchase from a brand that personalizes experiences, and that is nice. Okay, so how it works is AI driven product recommendations, they analyze the customer behavior, right, they offer personalized product suggestions. This increases your conversions because it’s presenting the relevant products right at the right time to those customers, and then that leads to enhanced customer satisfaction because you’re catering to their individual preferences.

So to do this you’re going to want to pick AI tools that analyze the browsing history, the purchasing patterns on your site, and provide those tailored recommendations. So some tools to look at for this are Insider, Braze, CleverTap. Salesforce also has a solution for this, it’s Commerce Cloud Einstein. If you get the slide deck after this, these links will go directly to those.

Okay, next is improving customer support, and I wouldn’t even say just customer support, sales as well. Sales and customer support and lead generation with AI chatbots, and this is kind of what we specialize. So nearly two-thirds of customers expect to check out online in four minutes or less, that was according to Capa, and honestly I think it’s less than that, you know, depending on motivation levels. If somebody’s highly motivated they will spend a longer time on site, but generally people aren’t. If they get what they want, they find what they want within, I’d say, two to three minutes to be honest with you, and if not, they’re out. And you’ve probably experienced this on your own, if you’ve been frustrated trying to find the right product or something on a website, or couldn’t get the answer to your question, you’re out of there, right? You don’t have time for that.

90% of users though will use chatbots when they’re offered, and what I found is it is the default, the first place people will go to if they have a question or they’re looking for an answer for something quickly. They’re like, is there chat, oh my gosh, is somebody actually going to answer it, right? Because if you do have a live chat it’s usually some sort of escalation protocol. And then again, 69%, can’t make this up, of customers were satisfied with their last chatbot interactions, right? They’re getting the answer to their question right away, they’re able to move on, make a purchase, and that is, okay, we don’t have to hear 15 seconds in Isis. All right.

So again, let’s look at some of the chatbot solutions. You’ve got Infinity AI Tools. Actually, let’s go back here because I want to talk a little bit about this. It’s not just customer service. Okay, yes, you’re getting 24/7 customer support because it’s answering those frequently asked questions, resolving all these issues, but the sales go up and the conversion rates go up because you’re helping your customers get to the products faster. And that’s what we found. You’ll see some of the case studies here in a little bit, that’s what we found, and some really interesting applications of where this is able to increase sales in ways that we didn’t think was possible. But just having those faster response times leads to higher customer satisfaction. There’s just this window of opportunity that you miss without having this scalable solution that can respond to somebody at that point where there’s some sort of friction to their conversion, and they just need an answer, they need a little bit more info.

And obviously there’s a massive reduction in the workload on your human support team, so it’s very expensive to have live agents and scale that if you have a high traffic site. If you’re something huge like Skims, right, by the Kardashians, like it’s just not possible. Okay, even their chatbot right now is instantly escalating everything, they’re like, we’re bogged down by load times and everything. It’s like, I don’t know why they’re not using an AI solution yet. So implement the chatbots, they handle the common questions, and then anything more complex you can direct to human agents. And one of my favorite things is actually making it to where the AI can answer anything, and people will ask some dumb questions, but that’s okay, we just make the AI really good so that way even if they ask a bad question it can give them a good answer. That’s kind of what my job is.

So obviously we have a solution for this at Infinity AI Tools, and we custom design these for e-commerce and we custom design them for each brand that we help. HubSpot has a solution but it is not e-commerce specific. And there’s another one called Chatfuel which is not site specific, so basically this is something that’s used for like Meta and X and other platforms, one of those messenger bots, so not really for your website. Tidio also has a solution, it does use Claude, if you’re familiar with Claude, instead of GPT. I personally think GPT works better in this role, and again we’re just using the AI and your data set, and I think GPT-4 does a better job than Claude with that.

Okay, next we’re going to talk about optimizing inventory management. We talked about supply chain issues, Overstock issues, out of stock issues. And the way it works is AI predicts demand and manages stock levels to reduce those stockouts and Overstock situations. So obviously the benefits of this is products are available when customers want them, which is big, right? They go to get it and it’s not there, you’re missing out on sales, opportunity cost, and it’s blown, you never get it back. Then obviously we know that some people are sitting heavy on a lot of inventory as well, so you don’t want to be adding more inventory, so you want to reduce the cost associated with that excess inventory as well. So there’s AI tools you can use for real-time inventory tracking and automate reordering based on demand forecast, and if you have a store you probably use these already. Magento is an open source option, you’re going to need some development to go with that, but Shopify, WooCommerce, a lot of these platforms have these built-in solutions already.

Let’s talk about boosting marketing efforts. So AI driven marketing uses data to basically send targeted email campaigns and social media ads. Again, if you’re already doing this, these are probably some integrated solutions that you’re already doing. If not, you need to be doing this. You can also, the benefit here is you’re increasing your marketing ROI because you’re targeting the right audience with personalized messages. Good marketing is all about right message to the right person at the right time, and that’s what a big part of this is, and AI can help with that in a big way. And obviously you’re saving time with automation as well, right, these are the proper things to be automating so you can save your team time, have them working on more important initiatives. So I would say automate your email marketing with AI to send those personalized offers and content based on the customer behavior.

So we’ve got some solutions for the content creation side. We actually do have full brand voice capability with these AI solutions, and I think we’ve got 14 different tools in a dashboard for anything from press releases to your website copy, landing pages, social media, you name it. And we put it head-to-head with in-house writers as well as far as brand voice goes, and we win every time, pretty much. No one has beat it, we haven’t had an in-house writer beat our AI copywriter. So I know a lot of people don’t like AI copywriters, but the customers do, so the customers prefer the AI copy over the old in-house copywriter, which doesn’t feel good, but it feels good for the company that makes more money right now. Omnisend, Klaviyo, here’s some other email providers that have these solutions built in, which is awesome. Claude is mind-blowing, yeah, Claude’s good for a lot of things. Again, I’m going to talk about a tool here in a little bit where you can play around with some of these different AI models and test, and I would say, hey, if you like Claude better, you’re getting the results you want, use that, use the one that works better for your application.

Let’s talk about some of the success stories that we’ve had. Okay, so Sir Lou Cigars, anybody here a cigar smoker? I am not, I do golf with some people who are cigar smokers, and they knew about C, which was cool, but I had no idea people spend so much money on cigars. So this was a huge win. As soon as we implemented some of these solutions we had just instant results. I think within the first seven days we saw some massive results. So here’s kind of what happened. So they had lower conversion rates, they had some high bounce rates, some lower site engagement. And so we implemented some personalized shopping experiences and assistance, and the AI chatbot engaging visitors right off the bat, and did some tailoring and customization for the customer support. So within about seven days we got these results: their conversion rates went up 18%, their customer engagement levels went way up, they went up 60%, and average order value went up 31%.

So people were actually purchasing directly from the AI sales assistant on the site, and there was a higher average order value, I think just because it was delivering those options much faster. And what we found is people would almost just, like, use it as if it was keywords, they would just type a cigar name, and then you’d get this full exhaustive response from the AI bot, all these different options for not just that cigar but related ones, different flavor profiles, and people kind of just went wild with it.

Okay, the next one is not an e-commerce store, and this is by design. It’s called New Smiles, and it’s a dentist office. So we do have some customers who are B2B and, you know, they’re more interested in lead gen and things like that. And so they wanted to try it, we said, hey, we’ll work with you, tell us what your goals are and we’re going to help you achieve them, we’ll customize this specifically for you. So I picked this one specifically because it’s just fascinating. We get just so much data from our customers, from behavioral data that we can then analyze and give to the brand. So their challenges were low appointment bookings for their dentists, and they do a lot of things, dentures and the crowns and all this stuff, and a high volume of customer demand. So it’s an older demographic and clientele for New Smiles, and so tons of customer support demand and load. So obviously even just the customer support alleviation would be a huge win for them.

So we gave the AI chatbots to engage the visitors, we customized it for online bill pay, and I’ll talk about that here in a second, and then did some more customization for the customer support specifically. And what happened was crazy. So more than 30% of people who used the AI requested to pay their bills. And what was happening was people couldn’t figure out how to pay their bills, even though it’s in the navigation, and there were some CRO issues going on, there could have been some high contrast, but there’s just this assumption that, hey, we put this thing on the website and people will click it or they’ll find it, it should be obvious, right? Well, it’s obvious to you because you work at the company and it’s your website, you know where that is, but to somebody who’s maybe not as technologically savvy, or it’s just too much trouble to find it on the website, they couldn’t figure out how to pay their bills. So now you’ve got one-third of the people who are using the chatbot paying their bills online, which is great. So they got a lot of revenue, also their appointments increased, and then customer satisfaction increased as well. But just a really cool example of, you just don’t know what you’re missing out on unless you implement a system like this.

So why should you care? Well, these case studies demonstrate just the tangible benefits. So you don’t know what’ll happen if you integrate AI into your e-commerce strategy. So we know it’s going to increase conversion rates, we know customer satisfaction is going to go up, and you can achieve similar results with your stores. So here’s how you can kind of get started doing this. Does this seem daunting? I know a lot of you said you’re already using AI, right? Like the pulp trolo, I’m a big meme guy, I like memes, I’m like ancient memes at this point. But are these solutions daunting to you, or is this just like, no, I’m in, like let’s do it, I’m ready to do this? Just kind of want to get a sense of where you guys are at, because even a few months ago people were like, ah, I’m not ready for this yet. But look, with the right steps you’re going to see the benefits immediately.

Okay, and we’re going to simplify the process. What data is required, how to prepare it, good, yeah, we’re going to talk about data. Okay, so here’s kind of a good starting place. You want to assess if you’re ready. So just do a quick Readiness check, and what this means is, what’s your current infrastructure, do you have the necessary data and technical capabilities, can you pinpoint areas where AI can provide immediate value? We already know some of them, customer support, marketing, inventory management. If you have these issues, that’s something to look at, so identify these key areas for impact.

Let’s talk about data availability. You need to ensure that you have quality data for the AI to work effectively, and I’ll talk about this a little bit because it’s really important. We had a trial Prospect, a trial customer, they’re like, your AI system is blowing it, dude, it’s telling people the wrong information and it’s really messing things up. And I’m like, wow, okay, let’s look at that, that’s a bummer, let’s check that out. It’s like, yeah, it’s telling everybody that we’re open on Fridays, or not open I should say, their website’s up but they don’t have support on Fridays for whatever reason, their support is only Monday through Thursday. So if you have a concern, don’t call them on Friday, yeah, we don’t answer questions on Fridays. And then it doesn’t matter anyway because the number’s wrong. I said, oh, interesting, okay, so we look at that. Well, turns out it was directly from their landing page. So I showed them the landing page on their website, I said, well, the contact information on this is directly from your contact information, it says Monday through Friday to call this number, so the number’s wrong and the days are wrong. I don’t know why they don’t help people on Fridays or Saturdays or Sundays, to be honest with you, but they don’t.

And so it’s a data issue, right? The AI is only as good as the data and only as accurate as the data. So if you have mistakes in your data, we’re going to talk about kind of hygiene here in a little bit, the AI doesn’t know, it only knows the data set that you give it, so making sure that that data is accurate is really important. And I know that there’s a lot of data changes, there’s a lot of site changes, store changes, especially during holiday season, right, you’ve got new promos, this promo sunsetting, this one’s, right, there’s a lot of movement, and a lot of different data. So it’s really critical that if you are using an AI system you’re on top of this, and you don’t have bad data in there. And I have lots of examples of this already with other customers.

Okay, next is technical infrastructure. So just assess if your current systems can support AI integrations, and most of these are very seamless integrations, right, a lot of ours are seamless integrations, and some others as well. And then just make sure your team’s on board. Some people are still a little nervous about it, like, hey, is this going to take my job, stuff like that, so just be cool with your team, explain what you’re doing. Ensure that they’re not going to lose their job, if anything this will help them keep their job. So if they can get really good at integrating these solutions, it’s, they’re going to get the wins, they’re going to hit those metrics that you’re looking for, and then they’re going to have more time to work on stuff they probably want to do. So we’ve got an AI Readiness checklist that you can go through kind of like to walk through some of these things, so if you want to look at that, again if you have the deck you can click the link, you can go to InfinityAITools.com/prepathon to get that.

So here’s some simple steps just to start integrating. Start with tools that are user-friendly and can integrate seamlessly with your existing systems. Don’t try to do something that’s like this whole new thing, we got to adopt this system, we got to teach everybody how to use it, like a new platform. Try to find stuff that just kind of can augment on seamlessly to what you already have, right? We’re all about zero effort for the customer part, like we do all the implementation, they don’t have to learn anything, they don’t have to do anything, it’s like we just give you our solutions. And then if you even want the data, we’ll give you all the data, you can customize, but we’ll do all that for you. It’s like, tell us what you want, we’ll do it for you, it’s implemented, it’s seamless, zero impact, you just get the results.

So you want solutions like that, look for those types of solutions. And then focus on one or two areas at a time, those high impact areas first. If it’s chatbots for customer service, if it’s personalized recommendations for product pages, whatever it is, focus on the low hanging fruit, the quick wins, visible impact, right? And that gets people pumped too, it gets your team excited about it, and then just gradually implement from there. Start small and scale.

Yeah, date changes can bring confusion, right? I mean, and you probably say this all the time, it’s like, oh, when did that promo end, when did this end? So it’s really important to make sure that all of that is accurate, especially even just from a customer support standpoint. You don’t want to frustrate your customers when they think a promo lasted longer than it did, or the pricing is wrong or something like that. Those inaccuracies really lead to low customer satisfaction and will confuse your AI.

So for just a smooth transition, again, ensure data quality. The AI thrives on the data, lives and dies by the data. Make sure your data is clean, it’s relevant, up to date. We do daily data syncs, so we will sync with our stores every day because we know, especially again during the season, a lot of promotions, a lot of things are coming and going and changes, new products are getting added, so we want to make sure the AI has that new knowledge. And then regularly review the performance of the tools. We even, they were talking about this with automation, right, this isn’t the fun stuff to do, maintaining, but you’ve got to do it. You want to review these things and make adjustments as needed, do that fine tuning regularly, clean and update your data sets, again making sure it’s good, have good data hygiene. And then use analytics to actually track the performance, right? A lot of times people just set things and forget them, it’s like, no, treat it as an employee, make sure it’s getting the results it’s supposed to get, and then identify those areas for improvement.

And then incorporate feedback that you get from both customers and team members. What is your team saying about these solutions? Is it working, is it not, is it getting the results you expected, how is it working forward facing with your customers? Do they like the solution, if they’re interacting with it, do they even notice the solution? A chatbot obviously will, but maybe they don’t, maybe it was better than the live agent, maybe they thought it was a live agent, maybe they like the product recommendations. Get that feedback. We always provide, we get tons of insight just from the conversations that we have, where you have hundreds of conversations with your customers a month now, and traditionally maybe you’re not getting that feedback from your customer support agents, but we review it and then provide that, and it’s valuable insights for the brands from these chatbots. So assess the Readiness, choose your tools, integrate and test them, and then monitor and optimize.

Okay, next I want to recommend some tools and some experts that are going to help you on this journey, provide some value and some insights for you. So one tool that I’ve been using for over a year now is called Magai, and it’s essentially an innovative tool for customizing AI models and then experimenting with new AI APIs, and this includes image and video generation. I know some people were asking about video generation and animation, so if that’s what you’re interested in, I would definitely recommend this. I will say it’s a little more advanced, that doesn’t mean that you can’t use this tool well as a beginner, it might feel a little bit overwhelming as a beginner, but even if you just dive in and start to slowly explore, I think it’ll help. Even if you’re just like, hey, I’m going to put in a prompt and see how different AI models respond and what kind of output I get, right, even just some very simple things like that is a good way to approach this. But if you’re really interested in fine-tuning AI solutions and creating things for your specific needs, then this is definitely a good tool I would recommend because it allows you to do that. It’s going to help you stay ahead of the curve as far as cutting edge AI developments and innovative tools because you’re going to get to use them in this kind of sandbox environment and just experiment with those new APIs. So can’t recommend it enough, go to magai.co for that.

Another good one, again this is a video creation one, so this would be more on your marketing side, is Syllaby, and it’s basically an AI driven tool for creating short form video content. We all know, if you’re doing social, it’s short form video content that’s crushing it, right, YouTube Shorts, TikToks, Reels, so that’s really what you need for your social media marketing. So this helps you streamline video creation, there’s some really cool stuff you can do in here, but it’s basically high quality video content quickly. They do these really cool challenges every month where they do these like faceless YouTube video challenges, and so that’s a good way to start this, is to just join up, do it for a month, do a faceless video challenge and get a feel for the content you can create. It’s really good for Instagram, TikTok, YouTube, this is where it counts, so if you’re reaching your customers on those platforms, this is definitely something to take a look at.

And then one expert I want to mention is Veronica Jeans. She’s a renowned expert in e-commerce marketing strategies, and she’s been helping businesses thrive online, and her areas of expertise are e-commerce strategies, effective marketing techniques, customer engagement and retention. So you can follow her, go to veronicajeans.com. And she was very gracious enough, she actually just came out with a new book called The Profit Playbook, and she very graciously created a holiday preparation checklist for you guys. So I told her I was doing this talk and she wanted to provide that for you guys, so if you scan the QR code there you can get that preparation checklist and check that out, and you can also again go to her website veronicajeans.com for that.

So again, I wanted to give you the free resources that I had available for you, so if you go to InfinityAITools.com/prepathon you can access both of these. There’s the AI Readiness checklist that you can go through, and then also there’s an advanced AI strategies guide, you know, once you start implementing these solutions, kind of like what those next steps are. So again you can click both these links, or you can just go to that URL and get both of these. So with that, are there any questions?

Brent: Colin, amazing presentation, love it. I’m going to jump over and help you manage the Q&A here for a quick second. If you have a question out there for Colin, ask it using the Q&A panel if you don’t mind, and we’ll just look at some of those latest questions there for you. While we’re pulling that up, let’s just give Colin a little bit of a round of applause, I’m seeing some amazing, excellent.

Colin Lepiscopo: Presentations, oh thank you, um, all good.

Brent: All right, so our first question here we’ll grab is from Maurice, this is from a few minutes ago. What are examples of data sets a web design agency would work with to train an AI model?

Colin Lepiscopo: Yeah, good question. So essentially you can take any data set that the brand has that they want to add, but the starting place is the site map, right? You need to start with the store and the site content, and that’s the best way, that’s the first data set really, that’s the first one. That’s why, if there’s any inaccuracies on that site map and on that data, you’re going to run into some real problems. So you want to start with that and train the AI on that, and then any additional data sources you want to add, right? So maybe they’ve got some customer support protocols, or they’ve got lots of other resources that maybe aren’t on their site, or some other information that they have internally, that’s the type of stuff you want to add. Sometimes it’s even just more technical things they have, like product tip sheets and things that go beyond what they would have as far as forward facing for the customer and marketing, but they’re more detailed questions that are relevant, but that information is just not available on the site. That’s some other data sources that can be made available, and those can be spreadsheets, PDFs, whatever.

Brent: Very cool. I feel like, from a personal standpoint, I’ve experienced AI in customer service a lot more and I’ve actually been very pleased as a customer, I think in most cases. Just the other day I got a refund from, I bought, I was trying to use Spotify for something and it didn’t work for what I wanted it for, and I just went on there and was like, I did it all through the AI, got it instantly done, something that would normally have to go to a live chat agent during a very specific time period, and they were able to process that all through their AI bot. So being able to actually do functions on the back end, not just answer documentation questions, but actually process returns and stuff, so that’s been pretty cool.

Colin Lepiscopo: Yeah, it’s huge, yeah, and that’s the next iteration of this for sure, is, you know, whatever API integrations it has and can do some of those backend things, or do some of those things that would normally get escalated if it is just a chat, is awesome. And then yeah, you’re stoked, right, you’re like, man, that was easy, I didn’t have to talk to somebody, I didn’t have to go through support emails and wait days, and it’s like you’re getting your answer, you’re getting your problem solved immediately, and it’s like, yeah, I’m good, that was great. And it’s not the norm, right, because the norm is, it doesn’t work, it escalates, I’m not getting my question answered, I got to go through this workflow, and it’s sending me to these pages or resources that are irrelevant to what my problem is, and you’ve got lower satisfaction. So yeah, great example.

Brent: I saw somebody said the first thing I say to an AI chatbot is Agent, which I do feel like is a little bit of a, I hear you on that one because I’ve done that many times myself, like you call in or you text in, you’re like, agent, agent. But I have to say, as I’ve seen AI improve and have those examples to share, like with Spotify who’s obviously doing it really well, I feel like I’m almost less inclined, like sometimes even with Amazon now I’m like, actually this is easier because I can get in, I don’t have to wait in line, I don’t have to wait in the queue. If you know how to prompt well, I think that actually helps you in dealing with customer support if it’s also being run through an AI agent as well.

Colin Lepiscopo: Yeah, absolutely, no, I think you’re right. Look, I’m the guy too, when I get on the automated call I’m like, operator, zero, pharmacist, take me to your leader, right, I like immediately, okay, so I totally get it. But that’s part of the challenge for me, is I can only manage a one-sided conversation, so the AI is only going to work as good as the prompt. And like I said, people are going to ask bad questions, or just not complete questions, or out of context questions, and so what you’re dealing with essentially is horrible prompts all the time, and it’s only as good as the customization that went into it from wherever you’re getting it from. So that’s what’s really critical, is we have to do our best to try to anticipate all these things, and the more customers we get and the more interactions and conversations we see, we just get better at building this. We have to be able to accommodate some poor prompting from these customers.

Brent: I love this question here from Maurice around, how can AI help streamline e-commerce logistics, especially as we’re going into Black Friday, Cyber Monday, all that kind of stuff?

Colin Lepiscopo: Yeah, well, we talked a little bit about that, and I think it’s the inventory management stuff, right, that’s really where it’s going to happen. Because what we’re dealing with is kind of these Predictive Analytics based on the customer behaviors, and so using that to help figure out your stock levels and stuff like that is where you’re going to kind of, that’ll have a trickle down effect, right? So you can’t affect the whole supply chain and all those logistics, but you can affect what’s on the end with the data that you have.

Brent: This question is coming in from Amir. How does AI contribute to better data-driven decision making, or maybe what are some additional ways that we can make this contribute to data-driven decision-making?

Colin Lepiscopo: Yeah, good question. So similarly to where you would get data if you looked anywhere, right, for the right data, like let’s say you’re going to look in Google Analytics, which is a good place to look for things, right, you’re curious and you have a question, you’re trying to figure something out. It’s the same way. If you’re trying to make a decision, or you’re getting curious about why customers are doing a certain thing, then you just look at that data set. So as far as the data that we’re getting from different AI systems, if it’s Predictive Analytics and it’s looking at the customers, like let’s say for personalization it’s looking at their browsing and it’s looking at their purchase history, and it’s taking all of this data about how this person has interacted with your brand, and then it’s giving you suggestions on how to message them and how to personalize that message, like they probably are going to be interested in these products, they’re probably going to want to know about this. So then you just already have it ready to go, and it can be automated to where it isn’t even a decision.

But what you can do is, paying attention to what the AI is telling you about your customers will help you learn more about them, right? That helps you make decisions, because this isn’t just like next campaign, this is, you want to get out ahead of your customers. If you’re a good brand you’re using all this data to go, what’s the next product line we need to unveil, I think they’re going to be interested in this. And like we’ve created some tools too for even further out, predicted, where we’ve got like an ideal customer avatar where we can be like, what do you want to buy next spring, and then we can help a fashion store plan out their next line, because we’re like, hey, we know that they want this light windbreaker, and based on your demographic and all this data, we kind of know the three or four items that they’re going to want next spring. So it’s like, now you know what your next product line is, so you can go actually further out and again test all these things, but it’s pretty powerful stuff. We’re just on the beginning cusp of some of this stuff.

Brent: So yeah, that’s cool. In your opinion, which companies are acing it when it comes to leveraging use of AI chatbots, again coming in from Maurice?

Colin Lepiscopo: Yeah, there are some solutions out there, I provided some in the deck that you can kind of look through. A lot of them aren’t catered specifically for e-commerce, some of them are catered specifically for e-commerce, and some of them are limited. So, you know, we’re on a lot of Shopify stores, we go head-to-head with the Shopify AI, and what we found is we’re getting better results than them. So you might have people that say, hey, we love the Shopify AI and stuff like that, you might say, hey, we love the integration with WooCommerce, right, so a lot of times people are just using the integrated solutions. We’ve spent a lot of time on the customization side, because we’re smaller, we can work one-on-one with our customers and customize these. So if there’s other companies like us that are doing that, I would say you kind of want, if you can get white glove, get white glove treatment. But with something like Shopify, they’ve got thousands of customers, so they’re just going to have to give you that cookie cutter solution that’s just not going to be tailored and kind of a bespoke solution for your brand.

Brent: This one’s coming in from Evans from chat. Any great software for animation apart from Blender?

Colin Lepiscopo: Oh, well there’s the video stuff. I would say, like, I don’t know about animation specifically, but video is definitely on the cusp. I would check out Magai because it’s a sandbox for all of these softwares, right, it’s got Leonardo, it’s got SDXL 1.0, like all the image-based ones that are now adding on video. And that’s the next kind of evolution, is the AI starts with an image and then they move to video, all of those are available. So I would say that’s a good place to go play with those, so not specifically a software, but like, you know, talk about the actual AI models themselves. And there’s lots of other ones that you can look at for images as well that will use lots of these models, but again Magai is kind of just, I would just go do a trial there, just check it out, that’s kind of a catch all for all of these different models. Whereas if you go to NightCafe or something, well, you get a bunch of image ones and you got to go over here and get like one video one, it’s just, this is the One-Stop shop for kind of all of these that makes it a lot easier, so kind of go check it out, and I’d say see which one you like, see which one is giving you the results that you’re looking for.

Brent: Very cool. Colin, this has been a fantastic presentation, I feel like it’s a very hot topic coming into this holiday season, so very timely for our Prepathon guests. Real quick, before we send you on your way, how has been your overall experience at Prepathon 2024?

Colin Lepiscopo: It’s been great, I loved being here, happy to present the information, and I hope you guys are just excited for this holiday season. I think it’s going to be a good season, and there’s just nothing stopping you from starting now, you got plenty of time. But it’s been great, it’s been awesome. I’m really passionate about helping store owners, so I was honored to get invited and get to present with you guys today, so it’s been great.

Brent: Awesome, well folks, let’s give Colin a nice round of virtual applause here at Prepathon, let’s give him some emoji and your favorite meme love. Colin is a meme aficionado, and so make sure we reward him properly. So if you have additional questions, Colin’s email’s up there, I saw some other chats like, hey, how can we get an Infinity AI Tools trial or stuff like that, so I’m not sure what you all are giving away, Colin, but if they email you they can probably find out more information.

Colin Lepiscopo: Yeah, yeah, email, go to the site, we actually have a 14 trial, so yeah, come check it out for sure. We’d love to, or if you guys have any questions, again, we make custom solutions and stuff like that as well.

Brent: Yeah, awesome, very cool. Well, we will hopefully see more updates from you and Infinity soon, so thanks again, Colin.

Colin Lepiscopo: Awesome, thanks guys.

Brent: All right, next up we have, we’re going to be doing a little bit of an Expo, a bonus Expo session. So for those of you that are looking to score some additional points on the leaderboard, we’re going to be spending the next 20 or so minutes diving into those booths, our sponsor and partner booths, the Expo booths here at Cloudways Prepathon 2024. So if you’re looking for more participation points on that leaderboard, if you want to win one of those contests, and we are going to be announcing winners imminently right after this Expo, we’re going to be announcing the Cloudways Prepathon 2024 winners. And so if you go visit the booths over the next 20 minutes, that’s going to give you an opportunity to rack up some additional points, and in those booths and at the Expo our partners are offering additional discounts and credits and offers and all that good stuff. So spend the next 20 minutes, check out those Expos, rack up some additional participation points in the leaderboard, and when we come back we will be announcing our contest winners for Cloudways Prepathon 2024. So we’ll be back in about 20 minutes, spend this time jumping into those Expos, saying hi to our partners, make them feel that Prepathon love that we have this whole time. And when we come back we’ll be announcing those winners for our Prepathon contest. So you guys have been so amazing, and we’ll see you here in a few minutes for those winners. All right, talk to you soon.