Zak (00:08.097)
Welcome to the D2C Revenue Rocket podcast. I'm Zak Cassady-Dorion, your host and CEO of ECD Digital Strategy. We are an e-commerce marketing agency specializing in paid ads, retention, and Shopify design and development. And today we are live at ShopTalk, and we have an amazing panel of guests here to talk to you about how e-com brands are leveraging AI for growth. Guys, welcome. Thank you for coming. I have Gina Potenza from Klaviyo.

Gina (00:34.274)
Thanks for having

Sebastian (00:34.862)
us.

Zak (00:37.346)
Dwayne Doshier from Shopify, Michael Epstein from Postpilot, and Sebastian Brown Glad from Clear.io. Can you guys go around and give a quick introduction, who you are, what you do, what your focus is?

Gina (00:48.32)
I'll kick us off. I think this is just, you're going to hear from me a lot first. Gina Potenza. So I'm with Klayvio on our partnerships team, but really focused on platform and AI and how we can work with our wonderful agency and tech partners to make sure that brands are getting the most out of the richness of data and customer relationships that live within Klayvio

Dwayne (01:09.346)
My name is Dwayne Doshier. I'm a partner growth architect at Shopify. I serve by helping our partners grow with Shopify. I spend a lot of time enabling our partners to better understand Shopify commercially and technically. I spend time with merchants doing the same, helping them understand our partner capabilities, and I'm really honored to be here.

Michael (01:28.767)
Michael Epstein, co-founder, co-CEO of Postpilot. We're a direct mail platform. We work with thousands of Shopify brands, Klaviyo brands to help them launch profitable direct mail campaigns, full funnel. So retention, retargeting, and new customer acquisition.

Sebastian (01:43.694)
I'm Seb Sebastian. I'm head of partnerships in the US for clear.io. My focus is on building our ecosystems for each one of our products and then the use cases that are created. I work with agencies to help develop interesting strategies, especially with the likes of Klayvio, how we integrate with Klayvio and how agencies can better leverage those integrations for their clients.

Zak (02:13.588)
Awesome. That was great. Thanks, guys. Yeah, so today the focus is really how companies are leveraging, how ecom companies are leveraging AI for growth. And the first question I'm to go through and have all you guys answer is, what does personalization at scale actually look like in practice? And how is AI making that possible? Dwayne, would you mind starting us off?

Dwayne (02:36.384)
Sure, I think personalization is, I mean it's, the holy grail of personalization is a single message to single buyer or single shopper, right? And in the, you know, let's call it five years ago, that was really difficult. Like you might be able to deliver 20 messages to 20 different segments, maybe, you know, 40 messages to 40 segments in different languages.

But now with AI, you can deliver, I think a lot of times, right message, right time, high amount of context with the prep, potentially what happened offline, what's happening online with Klaviyo, what's happening with other sources and bringing that all together. So doing that at scale could be doing it across digital touch points. It could be potentially doing that in offline touch points as well. And doing it not only at the number of customers you touch, but really how frequently you can touch those prospects and customers.

Zak (03:26.07)
Is Shopify spending a lot of time, energy and money on integrating with all these other softwares to be able to do that at scale?

Dwayne (03:35.502)
Well, I think so. I mean, in terms of some of the other software as you talk about, it's well known that we run the carton checkout for many of the agentic platforms. So we spend an incredible amount of time with the current agentic platforms, enabling them, of course, always thinking about the future as well. And in terms of some of the partners in the room, including Klaviyo, we spent many, many, many hours and years enabling first party solutions where not only Klaviyo is building integration, but Shopify might be co-building that integration.

And we don't just sit on that once, now, but three years ago that was built on REST, you know, now everything is built on GraphQL. We spend a lot of time deepening the integrations, especially with, you know, you know, partners such as Clay.

Zak (04:18.604)
Great, thanks. Michael, you're up.

Michael (04:21.101)
Well, I'm picking up on something that Dwayne said, which is the right message, right time. And that's really, think, from our perspective, what we've tried to pioneer and bring to market, which is you think about direct mail as a channel and you think it's like batch and blast. It's like, let's hit everybody in a zip code or let's hit everybody, know, females who are over 35 with a message and then, you know, a bunch of spreadsheets going around and then figure out what happened with it. What we've essentially built.

and what we've really been able to bring to market for brands is the ability to do one-to-one personalization of these cards. So every individual card that is going out is triggered in real time and personalized for that individual based on their purchase behavior, key demographics or characteristics about them. The card itself is personalized to them with the creative and the messaging that goes on it. That's, you know, I...

fully believe in that. I fully believe in what Dwayne said, a bright message, right time. That was sort of the thesis behind what we were trying to build. And so that's really what permeates everything that we do with direct mail as a channel for e-commerce brands today.

Zak (05:25.326)
I think you guys are nailing it too. Before I found Postpilot years ago, I was not into print because one, the attribution, was just tough. And then the personalization wasn't there. But the way that you guys tie in with Shopify, tie in with Klaviyo, it's awesome. Now leveraging AI, I think there's some awesome things you can do with that to drive meaningful revenue. So.

Michael (05:49.005)
Absolutely.

Sebastian (05:50.67)
I guess it's basically the same idea, different wording, I would say responding to behavior in real time. So I think at scale, what we have done with Boost is that we're looking at intent signals through discovery and search and filtering as really, really important key data points that will inform

any shoppers experience as they go along through our recommendation widgets. And as the shoppers going through their customer journey, they're being spoken to by how they're behaving on the site, which definitely would inform how that data is being captured. And then what we can do with that data, obviously, is porting it into Clavio for smarter marketing and for agencies such as yourself to do consulting around what that data does.

Zak (06:47.478)
I love it. That's great. If there's anyone that doesn't know what it is, would you mind giving a little background on or little more details on what boost is exactly?

Sebastian (06:55.15)
Yeah, Boost is an AI search and product discovery platform. Basically it's a search and filter platform, usually for like customers at around a million to a hundred million GMB. So if as you're moving out of the native Shopify search experience and you're looking for a little bit more robust features and

high products view counts and complex catalogs. That's where we can come in. We're built within Shopify. And then we also have a really strong personalization feature with our recommendation widgets. And those are showing up in Cart 2. It's basically personalization and search and filtering.

Zak (07:41.23)
I had a light bulb going off in my head here. An integration with what you guys are doing in post-pilot. So if someone goes and searches a product two or three times, they don't purchase it, then that data gets sent over to post-pilot and they get a direct mail postcard in the mail trying to get them to finish that purchase.

Michael (08:02.798)
We just released a new open API that allows us to take events from lots of tools like yours. And then we also work directly with Klayvio. And so anything that feeds into Klayvio can then be used as a trigger for a direct mail campaign as well.

Sebastian (08:17.934)
Yeah, and we capture that data for sure. Yeah, I'm here. All right.

Michael (08:22.463)
right now.

Zak (08:23.678)
And 3 % commission is all I ask. I love it. Gina, and I'll repeat the question because it's been a little while. Just so it's fresh. Thank you. What does personalization at scale actually look like in practice? And how is AI making that possible?

Gina (08:37.665)
Yeah, right place, right person, right time. I couldn't agree more. I think that, thank you all. You all just reinforced about saying the word CRM, what role Klaviyo plays, right? And how important it is for us to integrate with all of our partners. I will say that the piece that gets very overwhelming when I speak to marketers at brands and at our customers is that it's hard.

Right, that idea of being able to do it at scale one, how do you trust the AI that it's actually going to deliver it and that it's going to be personalized and it's not going to feel the same as all your competitors, right? They are likely using the same AI models. And also knowing that, which one of these is going to drive results if I'm doing the monitoring, if I'm being the human in the loop. so actually thinking about how can you use the right data to provide as much context as

possible so that that personalization is actually driven by all these different touch points, that's because of interaction with search, POS in person, or search or actually scanning the QR code on a piece of digital mail. That right place, right time is actually the right channel, right message, which AI is making a lot easier because you can test it at scale. so some of it is

And it's, of course, everything is going to be vibe now, right? But some of it is actually the idea of being able to go zero to 0.7 with vibe marketing, right? And effectively actually being able to use a composer within Klaviyo to go and see, if I were to do this, how far can I, how do I shake this out? Actually lowering that bar for marketers to be able to go use that data. And I often talk about this where it's really easy to be the dragon sitting on your hoard.

And that's all your data and all that beautiful ways that you can interact one-to-one with a consumer and with your purchaser and why you know Gina is buying differently than you three and Actually AI lowers that bar and makes it more accessible to marketers where you're able to do it and it's still right It's not segments of one person yet But we'll get there because you're able to trigger off that level of detail from all the context across the customer

Dwayne (10:57.55)
journey. Yeah, I love all that. I'd offer two if I may. You said something in the beginning, like how do you trust it? I'd say you don't. I think one of the things I might offer to our merchants, even to our partners is in the early days, which is all every day right now, check on it, check on your agents. And this is not a set and forget it type of thing. Right. And so while you might be seeing efficiency gains in your company from the use of AI tools,

You should still be checking on it daily potentially, right? And I see other partners who have BDR motions and they might have reduced some headcount in BDR and now they have 20 agents running around. They need to be checking on that. And it's not performing the way I want it? Is it still data syncing into some of these basics? So this is, we're not operating in a trust but verify type of situation. We're operating in a check.

potentially verified type situations. So that'd be kind of our takeaway from what merchants could be doing with AI.

Gina (12:02.158)
Actually, sorry, Zak, I think that to plug actually, Zak and the team, this is the power of working with experts in all of our platforms where we're all the tech tools that can tell you. And as a marketer, that feeling of overwhelm is when having experts and agency partners who actually are really deeply embedded and know what does it even mean? What is the checklist? What are you verifying against? So much of that is the processes that are actually built into what.

You all have those for your clients.

Zak (12:32.998)
Yeah, it's true. We don't do everything under the sun, but we do very specific things. And the learnings that we have from repeating them over and over over over again with different clients is you see something that happens on one and then you know where it doesn't matter what that other product is, something similar is gonna happen there. So you need to watch out and make sure it doesn't happen. Something else that you said too I liked was you don't go from, you didn't say zero to one, you said, know, vibe coding and AI in general.

gets you from zero to point seven. Because that's not the trust but verify but that's the verify and make sure it's there get it you know, shorten your time to go live, but then make sure that everything is in place and accurate. Great, that was great. Gina, you guys at Klayvio have had some major, major changes, which are all awesome. In the last like

nine months. And the changes that you guys have made have actually forced us to change our roles in our company. We used to call the people that manage Klaviyo accounts, email marketing managers, that that role is gone, it does not exist at ECD anymore. They're not, they are now called lifecycle marketing managers, which takes into account all the other retention stuff that we do. But also, it takes into account

all of the other skews that you guys have come out with, all your case service lines. So that leads me to my next question. Klaviyo is rolling out a lot with case service. And how is AI actually being used to drive meaningful, incremental, trackable revenue?

Gina (14:16.758)
Yeah, I think that to the point of life cycle, to the point of customer relationship management, to the point just how much data is in Clayview, it's that, you know, I think for a long time, for better or for worse, we haven't thought about the post purchase experience in terms of actually support being tied into being a revenue engine, right? And actually what about pre-purchase support? And so we're widening the aperture on the idea that fundamentally

Gina is still your buyer and it doesn't matter where in the web of a customer journey. It's not a neat circle. It's not a funnel. These buyers have a lot of different touch points. so, Klaviyo service is almost the last piece in the puzzle to unlock the fact that you are able to consolidate all of those moving pieces of your relationships with your buyers to meet them where they are personalization at scale. What that actually ends up showing up as is I think that

It can feel, there's a lot of AI hype and it can feel really intimidating again in this idea that like, it's going to take over the world. But actually some of it is making sure that, Hey, if you know that you are selling a pair of leggings and the most common question that gets chatted in about in the most common return reason is care or fabric questions. How are you making sure that AI is actually preempting whether that's an abandoned browse and an abandoned cart?

you're actually prompting in that follow-up series, wherever you're contacting, right? Having an AI response ready, having a prompt saying, do you have any questions about texture material sizing? And then having AI without having a human be able to respond in a switch, right? Whether that's also then followed up with product recommendations based on past purchase bait or we have all of this data that again is training the AI, but.

you're able to scale and drive revenue because you're consolidating the entire vision of who a person is, not just getting them to that first purchase, right? It's not enough.

Zak (16:18.798)
And it's not just driving the revenue, but you're able to track it now. Because customer service has always driven revenue. But historically, it's not something that we have focused on because if it's phone calls, you can't really track the revenue. But now what we're seeing with our clients where we're using case service is that you're actually able to see the incremental revenue that is driving because all that data is all in one place. It's similar to what got us into direct mail with being able to see the

trackable revenue is the same thing that's getting us into K service. It's like the most expensive customer is always that first customer. And what keeps people coming back is great experiences and great products, great customer service. And so now you can have that customer service, you can track the revenue. I love what you guys are doing.

Gina (17:04.781)
Thank you. We're excited about it.

Zak (17:07.342)
I'm a little excited about it too. Dwayne, how is AI changing the way brands activate their commerce data for marketing?

Dwayne (17:16.642)
Gosh, really, think with AI, so much data that was previously disconnected is now being connected to Klaviyo. It's being connected to Shopify storefronts. It's being connected to post-purchase partners. All these different data sources.

put offline activity, activity on other sites that maybe aren't a Shopify site, being brought to personalize the experience. I think then in terms of marketing it as well, it's like, do we have more data points that are capturing signals, intent to buy, intent to churn, and being able to market to that?

In terms of what's going on with Shopify and Klaviyo, it's a very, very tight integration. And so bidirectional data sharing, things that are happening potentially in the service path, unrelated to the Shopify storefront can be brought to the Shopify storefront to influence purchase behavior, to influence churn reduction type of behavior. Similarly, what's happening on the Shopify storefront in terms of browsing can be brought back to Klaviyo, brought back to other partners.

market to them, And so, you know, we've often, we've always known, hey, you can retarget with ads or we can kind of retarget with search behavior. But now with AI, especially by thinking about, you know, AI think is very, very good at non-obvious pattern matching, right? And so how do we build, you know, on the fly segments of cohorts that did some crazy non-obvious behavior on the site and then market to them. And then if we discover that they are buying at a high,

rate, let's run a whole bunch more experiments like that, at scale, hundreds of them at a time or thousands of them at a time online, offline, and get after that. The ability for marketers to access data many times without involving the IT department is just unparalleled today, even compared to what it was a month ago, much less a year ago, five years ago.

Zak (19:21.09)
Yeah, that last piece I was waiting for you to say that the ability for marketers to access that data because you guys have always had an enormous amount of data is awesome. But it's it's you know, it's so much data that you know, historically it would take a whole team of people to be able to go through all day long and analyze that data. Now just with your guys' internal AI chatbot, you're a marketer able to get access to awesome data answers to questions that they've used to dig through reports for.

Dwayne (19:49.12)
Yeah, if you give me a moment to talk about Shopify, now, our sidekick tool does enable marketers, executives, really anyone on the platform to ask questions in natural language and get an answer. And so as I have dinners with some of our C-suite executives or our merchants and I talk about sidekick, I said, you can just go in here and ask the question, what was our fastest moving product yesterday? It'll return an answer to you. They're kind of blown away by that. And then they start to ask questions, well,

I have to wait for the Google Analytics report to refresh. And generally the answer is no. We democratize that in such a way that anyone at your company with the appropriate permissions can ask questions, get back some insights, and not only what happened, but in many cases what should be done about it. And so again, just like...that's all happening through the power of AI. There's not a, you know, not a team of data entry people somewhere inside Shopify doing that. That's the magic of AI. It's accessible not only to our merchants, it's accessible to partners as well. The merchants give them, you know, access to do that. And we encourage that.

Zak (20:54.21)
Yeah, it's something I'm loving as an agency owner. A lot of our clients, historically, would send us these one off questions and these one off ideas. And you know, we'd have to go and run them down and get back to them. But now they're able to run down these ideas themselves. And they're able to come to us with some other ideas that could be actionable. So it's great for us. It's great for them. It's great for the customers. All right, Michael, I am excited for this question, because it's it's about a new

a new product that you guys are doing that I am loving using with our clients. So I want to hear you dive into it a little bit. How is Postpilot using AI to help brands with their acquisition?

Michael (21:36.162)
funny you should ask that. Well, breaking news everybody. Just this morning at Shop Talk, we announced our new acquisition AI technology, is the next generation prospecting tool for direct mail for Shopify brands and e-commerce brands. And to the point that everyone I think is making here, it is really about connecting all of these disparate data points that we have.

billions of attributes across hundreds of millions of US consumers, essentially the entire US consumer data set. How do you make those connections to understand who is, like, not only looks like your best customer, but is actually likely to buy or is ready to buy at any given point in time. And so that was really the big innovation that we released today, which is the culmination of things that we've been working on with partners like Zak and have been testing this and getting great results for a lot of brands is,

Zak (22:13.531)
is

Michael (22:34.326)
Okay, using first party data through our data integration with Shopify, taking all those signals, using AI to make those connections between all these different attributes that we have, again, across the entire US consumer data set, and understand who is likely to buy right now at this given moment in time. And then you can run that campaign like you would a meta campaign. So it's not the traditional, know, direct mail batch and blast again of like, you know,

set this up for six months from now, throw out, know, mail a bunch of catalogs, wait a few months and figure out what happened. We're literally dripping out cards. You set a daily budget just like you would with a meta campaign. You're finding who are the best people that satisfy that budget at this moment in time, dripping out cards to them, and then just repeating it every single day. So you're managing budgets, you're managing performance like you would a meta campaign. And all these signals are always going to be fresh.

because of the fact that we are looking at this using AI to pick up on those signals and make those determinations on a daily basis, send it out on a daily basis, and then use the feedback that we're getting because again, our Shopify integration allows us to see the conversions that are coming in, use that to continue to refine and optimize those algorithms so that we're just continuously improving over time.

Zak (23:52.896)
I love it. And how are you seeing brands testing this out and having success with it? I mean, are they taking portions of their meta acquisition spend and bringing it over to Postpilot for that acquisition?

Michael (24:06.19)
Yeah, usually they're taking some portion of their acquisition budget, they either have, they're either taking it from an existing channel where maybe the next marginal dollar that they're investing in that channel is not gonna be returning as high as they would. Again, we generally all know that there's a point of diminishing returns with ad channels. And so, you know, what you got with the first 10 or $100,000 of spend on a channel is not gonna be the same performance you get on the next $100,000 of spend. So where do I...

direct my next incremental dollar of ad spend budget. And we say, this is a place that you should be experimenting with because not only can we deliver performance that is going to be commensurate with what you're getting across other channels, but you're diversifying your, your spent, right? Like, let's be honest. Like we all see what happens when you put all your eggs in, know, the meta basket or

You know, just the volatility that we see across different channels at different points in time. The fact that they're all subject to auction dynamics. So they get more competitive and more difficult over time. Whereas direct mail is a fixed cost and you're not paying more depending on who you're targeting or what time of year it is. So the long winded answer to your question is they're typically taking some portion of the spend or they have a discretionary or sort of exploratory budget that they'll use for it. I mean, basically if you tell your CFO, you know, I,

If I can get $20,000 or $50,000 or $100,000 and return you $300,000 for that, can we test that out? Brands are typically eager to find more channels that can do that for them.

Zak (25:44.522)
In your integrations with everybody here in your future one with a boost, I can see like a great great compliment to that. So you do an acquisition campaign, send out that postcard, someone goes to the website, they don't purchase right there, but they sign up for an email or SMS with Klaviyo. They get the flow there.

Maybe they still don't purchase, but they go to the website, they do a few searches, and then this future integration with Boost, you're gonna be able to then send another postcard out and say, hey, you left these searches in your cart, why don't you come back and complete your purchase.

Michael (26:23.822)
That's exactly right. Yeah, it's all a signal, right? And it's to the point that we were all making earlier, what are those signals telling us about how to interact with this particular person at this particular point in time? And to your point, how I know you do it with the brands that you work with, with Klaviyo and us, we're not saying replace your email with a postcard. We always say go to email first. That's going to be the best, cheapest option to get in front of a customer. But we all know that

Only a certain percentage of people are actively engaging with email at any given time and you see a ton of success with your clients by saying, okay, they've gone through this particular email sequence, they still haven't taken the action that we want them to take. How do we then trigger a card to go out at that time to get in front of them and try and encourage them to take that action before we risk losing them?

Zak (27:14.552)
Want to see where your e-commerce brand is leaving money on the table? Book a free revenue audit with the team at ECD Digital Strategy and get a clear plan to increase revenue. Go to clavio.com forward slash ECD to schedule yours today.

Zak (27:30.478)
Seb, where are you seeing brands leaving revenue on the table today?

Sebastian (27:36.398)
I have a two-pronged answer for that. So for us, it's at the discovery stage. I think that there's a lot of really important data that's being captured at the retention stage, at the card stage. But really, we're seeing that intent through search, filtering are those very, very early signals.

of when a shopper is actually very interested in what they're looking for. And when you can capture that data earlier and help draw them through that experience, that is, and if a merchant is focusing on that and what that data is like, then you're getting ahead of so much of the experience through the customer journey at that initial touch point.

I think as a shopper, I'm most interested in the thing that I'm looking for. When I'm, I have that first question, whether it's either in Google or on a site looking for something, that's where the most powerful emotion is, I think. And then it tends to drop off as I can't look for them, can't find the thing or it's not the right color or whatever the case may be. So then the other thing is trust. So, intent plus trust I think is really where.

that the twain met. So like we have a reviews product reviews.io and so that integration is really, really important with what we're talking about because as you're emoting while searching a product shows up with high ratings and reviews and all of a sudden you feel like held, you know, like actually, yeah, I'm going to buy that right now and maybe it's discounted.

or maybe whatever the case may be. And so I think getting ahead of that from that very first touch point is really where brands could be making more money. And then we have attribute data coming from reviews. So that becomes really important in the post purchase experience through Klaviyo. So if you do like a boost in reviews integration and that gets integrated into Klaviyo, then you have so much profile data that you can then fill out.

Sebastian (30:01.336)
to understand your customer, who's shopping what, why, and if you are leveraging AI tools like Sidekick or anything in Klaviyo, like the new product that you're talking about, then you can help that customer out in a number of different ways and understand what they're doing and predict what they're gonna do next. So that's my first answer. The second is...

I feel like brands should always be integrating their tech stack. I'm always really surprised when I see a brand and they have like this really luscious best of breed tech stack and like none of them are connected and no one's telling them to connect them or how to do that. And so, you know, we're working with data silos.

And I think we're just in an age in e-commerce where it's no longer about one single vendor, it's about the ecosystem and how each one of these vendors augments each other's solutions. And I love how you're coming up with these cases on the fly. Like that's so cool. And like leaning into agencies to really help understand what the ecosystem is looking for and like really encouraging brands to be like, no, you actually have to

All these things that you have, have integrations, just connect them. Because then all of a sudden you are like, you're soaring. I know, like you have, we put so much time and energy into developing these use cases and integrations and we're really trying to prove them out. But in reality, like if you just start connecting these integrations, then all of a sudden you're going to be able to see return on investment very quickly with fine platforms and stuff.

Zak (31:49.582)
I love that. I mean, there's so much data out there and it doesn't do a lot of good when all that data is sitting in its own silo. I mean, Shopify massive amounts of data, Klaviyo massive amounts of All you guys have massive amounts of data and it's great in and of itself, but once that data can talk to each other and it is leveraging each other, then that's where the magic happens. You said something too in the beginning that sort of played into what we were talking about before we went live about...

Like the magic happens in the discovery. That's like when people's dopamine is the highest. We were talking before we went live about how we don't see agents, or Duane was saying, I don't see agents taking over shopping. That's always gonna exist, but it's gonna be there for discovery. So AI will be there to help make that discovery and that decision-making process easier, but it's gonna need that person to be there to take it across the finish line.

Dwayne (32:42.574)
Yeah, think offline I was talking a little bit about my experience and then some of the brands that I'm interacting with, the partners I'm interacting with are seeing really good traction with using agentic commerce for discovery, top of funnel. So back to your comment about trust, it's super important for brands to have trustworthy content being fed to agentic channels. Like, know, not only product, attribute, size, color, you know, price, but how are people using it in the real world?

Yeah, how's it being reviewed? How safe is it? You know, is it safe for my child? Is it safe for my teenager? sorts of things. I really like what you said about your missed opportunities, you know, kind of top of a little discovery right there. You know, I also think about trust and, you know, the era of disconnected systems, my travel ticketing systems over here, my commerce systems over here, my manufacturing systems over here. And, you know, nobody, you know, you call the call center and the person doesn't know what's happening to these other channels. That year it could be effectively over if the tech stack is better integrated. My trouble taking system, my mail records, my search, whatever the case is. And then again, how much of that can be fed to top of funnel discovery to help my brand be discovered, make it positioned as more relevant than the brand who isn't providing these types of stuff, these types of information. then forcing that through to a purchase.

I think we're in that era now for the brands that want to invest in that and the cost of investment is fractional compared to what it was a year ago, much less fractional compared to what it was five years ago or 10 years ago. We're not living in an era where it's like with enough time and budget, I can do these things. We're living in an era where it's like with experimental budgets, we could prove this case out and do it pretty quickly.

Zak (34:33.293)
All right, well, I'm being signaled that we're coming up on time and I still have like six, seven, eight questions. We're not gonna get through all of them. But this is the closing question and answer this question, not specifically just on your software, but like in general, I think. There is so much happening right now and it's happening so quickly. And the more you learn, the more you realize

just how much there is. And it can be overwhelming at times just because of like, for me, the more I learn, the more I'm like, my goodness, how did I not know this a week ago, a month ago, who knew about this a month ago? And I didn't like it, so what else is there out there? But I think you can't do everything. You can only do very few things. And so your guys advice, what's one thing a brand can implement this week in relation to

everything that we've talked about.

Michael (35:33.248)
I'll start with AI in general. I think get on cloud, get on co-work, connect your tools, and set up a handful of scheduled tasks that proactively surface information to you based on, it could be anything from look at my calendar,

go through all the history in my CRM of every correspondence we've had and give me like the breakdown of everybody, know, the history of everybody I'm meeting with today and key things to talk about and go online and search for five cool icebreakers about it or something, whatever it might be. I think the key is you just got to get in there and start using it. I hear about that, you there's a question I'm asking everybody I come into contact with, other founders and so on. How do you get the team, how do you get your company, how do you get the team really activated here?

There's tons of different ideas and answers. Not all of us, I'm on, you know, X 22 hours a day. feels like at this point, feeling like I'm falling behind in every passing second because of how much new stuff is coming out every day. But I think realistically, it's your point, you can't do everything at once. like start, pick a tool, start doing like core use cases that you have today, and then start thinking of like, what are the things that I could do?

that would make my life a little bit easier today and just start doing them. And one, you know, just one new thing a day starts to compound really, really fast. And I think that's the part people tend to get intimidated by. Where do I start? Where do I get information about what to do? Who's gonna hand me the use case and just tell me like, go do that? I just don't think that that works particularly well. I think you have to get in there, start connecting stuff.

and then start just playing around with it and figuring stuff out and you, stuff's gonna start clicking to you. And then it's gonna become second nature. And then it's gonna be like, you truly start thinking about AI first at that point. You stop going to Google, you stop waiting for somebody to tell you what to do. You just say, oh, I'm just gonna ask AI how to do that. And then it becomes second nature. And then all of a sudden you're off to the races and you're that 10X employee that everybody's talking about now. How do you get people to become that? Like you just gotta get in there and start doing it. And it's so easy to connect your tools now.

Michael (37:54.552)
that there's not really an excuse to not be experimenting with that.

Zak (37:59.182)
Yeah, I like that a lot, one thing at a time. A couple of weeks ago, my one thing was a Sunday was rainy or it was snowy. I live in upstate New York. And one thing that I wanted to do that day was do a Klayvio audit using Claude. And I had this use case I was going to do and I went and I started doing it. You know, I fed in a whole bunch of our past audits of Klayvio accounts and

gave us a prompt, what I was looking for, connected Claude to Klayvio’s MCP. And I did what I was looking for, but then while I was doing that, I saw like six other things. was like, oh, I can connect this to LevelBull and create a dashboard. But nobody could have told me, okay, yeah, you do this. Or if they did tell me, I would have been like, I have no idea how to do it. Yeah, so I like that one thing at a time.

Sebastian (38:50.638)
Yeah, mean, I guess you kind of nailed it. And it kind of echoes the sentiment of like, connect all your tools and put all of that data into one place like Klaviyo. Start there and see what you're getting. Leverage the AI from Klaviyo to understand quickly what you have.

And then that will give you a new baseline for what you need to go get or what you need to get rid of. I was going to say one other thing, but I think that's really it.

Zak (39:27.778)
Awesome, I like it. Yeah, and again, it's pulling all your data together and then figuring out how to consolidate it and what you can get rid of.

Gina (39:35.438)
Okay, I have a spicy take that is actually a little bit analogous to start to the point that it is very overwhelming and there is all this data and like, if I could get anything I want out of Shopify, what would I even ask for? I would say actually take the time to map out your customer's journey. When you think about growth,

and map out what is the ideal, okay. The ideal flow state is that someone finds you, where do they find you? They buy the thing and then they become a customer forever. Fine, that's not interesting, right? Take it one level deeper, but then actually use your LML of choice, LML, I don't even know how to talk anymore about AI, but use it to actually ask it to interview you.

to figure out your blind spots about the customer journey. And that's actually going to tell it, these are the tools that I have. What should I go look for? What should I ask for in Shopify? Start filling in those links. And what's really important is if you are using a cloud, you can then ask it to create that as a doc. you don't have to, literally you don't have to know what a rag is to use the document of your ideal customer journey.

with the data from all of your systems as a rag for your projects, then you have this really deep nuance context that your competitors do not have and have not thought about. So everything that you do from there, you can keep building on it to actually have outputs that are really truly about you and the brand. And to some point, you'll figure out where your gaps are. You don't have that information. Let's go figure out how you can go get it, how you can build that experience.

Ultimately, I'm very aligned with you that people aren't going to stop shopping. We're evolved monkeys, right? It's not like the robots are going to totally take over for us. We all want the dopamine of shopping and that brain experience. How can you use that really deeply human edge to make one-to-one personalization actually meet your buyers where they are and use the human brain that we can't rely on AI to replace?

Dwayne (41:48.27)
I love that. I love all these comments. I probably have a two-party answer, one like off platform and one on platform for Shopify. But I really like what Michael said is, get started. Spend some time with Claude. Start asking questions. Potentially start asking questions about the other parkers in this room and their capabilities. Ask questions about your readiness to potentially take on a print mail program, take on search optimization, take on customer relationship management or ticketing or all the different.

capabilities that Klaviyo has. I think just getting started with Claude is the way to do that. For those brands in the room who are questioning how to get started, I would encourage you to read Toby, our CEO's memo, where he tasked the company to get started and says this will be in your performance video. Sure enough, it was six weeks later, right? So there's a little bit of care and stick. I'd say then on platform for Shopify to close out, play with our AI tools. We have Shopify Magic, which is a

content creation AI tool that can create compelling photography. Many of our merchants use it for SEO and it can be doing hundreds of rows at a time. It's a real efficiency gain there. And then in terms of sidekick, just go and ask questions. Ask questions about your Shopify storefront. Ask questions about what should be done and just play with the on-platform tools. And then maybe...

do both, right? Ask Claude for what competitors are doing or what partners should be considered and then ask Sidekick about it or ask Sidekick for some gaps and then see what Claude can solve and just get him into that virtuosity cycle of experimentation.

Zak (43:29.462)
Yeah, yeah, for me, I think the biggest thing is that, you know, we just need to realize we don't know what we don't know. And the only way to find that out is one step at a time. Every question, every time I ask a sidekick a question, I think of another question to ask and then another one, then another one. And then I start thinking of these use cases of what I can do with all this data. Yeah, so it's one step at a time.

Don't depend too much on what you're getting from the AI. Remember that it's real people in real life. This has been great. Great insights. Been great talking with you guys. And thank you for coming here today to be on the podcast.

Sebastian (44:09.954)
This was

Zak (44:11.31)
Before we wrap up, if you're serious about growing your e-commerce brand, go grab a free revenue audit. We find the revenue you're leaving on the table. We'll analyze your store, marketing, retention channels to show you exactly where the biggest revenue opportunities are. It's completely free and you'll walk away with a clear growth plan. Just go to klayvio.com/ECD and book your audit today.