Your Ads Aren’t Broken. Your Data Is. (Here’s the Fix)

Steve Hutt:
Welcome back to eCommerce Fastlane. I’m your host, Steve Hutt.

Here’s something that’s probably happening on your store right now and you don’t even know it. Your Meta Ads Manager might be telling you that you got 47 conversions this month, Google Analytics is saying 31, and Shopify Analytics is saying you have 52. You’ve got three platforms with three different numbers, and you’re making all these spend decisions based on data you can’t completely trust.

I remember going down this journey when I was inside Shopify. A lot of people would ask me, “Why is it that Shopify Analytics shows something different than what Google Analytics is showing? Why is that? And what is the source of truth?”

Over the years, there’s been some interesting tension around this. The answer was often, “Well, I guess you should probably look at GA as your source of truth if you’re really going to have one,” but even that isn’t completely accurate. So it becomes this ongoing journey.

Today, I want to dig into something I believe is at the root of a lot of these marketing problems. I’d argue that in many cases, it’s actually an infrastructure problem. That’s where I think my guest is really going to help us.

His name is Sean Larkin. Sean is the founder of a company called Fueled. They’re a first-party data platform built specifically for Shopify that captures every customer event and pipes out clean, standardized data you can send to your ad platforms, analytics tools, and email marketing.

It’s really interesting because they use server-side tracking instead of what I’d call the more fragile browser pixels, which I’m sure we’ll get into. The reason I’m having Sean on today is that Fueled was acquired by a company called Opensend, which has been known for identifying anonymous website visitors.

Sean is now the Chief Product Officer at Opensend, and he’s combining these platforms to do something pretty powerful: connecting who your visitors are with what they’re actually doing on your website. It’s a really interesting one-two punch.

So today, I’m hoping we’ll unpack what’s broken in the data, why that’s likely costing you a lot of money in the form of incomplete data, and how to get your marketing dollars working a lot harder.

So, hey Sean, welcome to the show.

Sean Larkin:
Thanks, Steve. It’s really great to be here. I appreciate it.

Steve Hutt:
My pleasure. Let’s start with what Fueled is as a product. Explain what it does in plain language—what problem does it solve, and why should a Shopify brand really care about having a clean data pipeline?

Sean Larkin:
Yeah, for sure. In the shortest—and slightly nerdy—terms, Fueled is what we call a customer data platform for ecommerce.

Customer data platforms are essentially a set of pipes—really intelligent pipes—that help you capture behavioral events: people clicking through your website, making purchases, signing up for subscriptions, joining your newsletter, things like that. We capture all those events and then send them to various destinations. Those destinations might be advertising platforms, email marketing platforms, SMS platforms, data warehouses—you name it.

That’s what we are in the nerdy sense.

In a more practical sense for everyday marketers, we help brands on Shopify manage their pixels and manage the signal they’re sending to their advertising platforms like Facebook and TikTok, as well as their email platforms like Klaviyo.

We see it as two layers. There’s the visible layer for marketers: your creative, your campaigns, your offers. Then there’s the invisible layer of data that’s powering all the algorithms on these advertising platforms.

What we do is help brands optimize that invisible layer so they can trust their data. They’re not freaking out, they’re not getting emails from the CEO at midnight saying, “I don’t understand what I’m seeing in Google Analytics,” or “Why is Facebook saying X, Y, and Z?” We handle all that data moving back and forth between platforms. We optimize it so brands can focus on creative and have a high level of confidence that they’re sending the right signal.

Steve Hutt:
Yeah, it’s interesting. You’ve got brands making ad spend decisions based on data they really can’t trust, but they’re still making those decisions—often without having your platform connected.

Can you walk me through what that actually looks like day to day? What’s breaking, and why don’t merchants realize there’s this disparity between what’s actually happening live on their site and what the ad platforms are showing? And then they’re spending money based on, “Are these campaigns actually working?” I’m curious about your feedback on that.

Sean Larkin:
Yeah, for sure. What’s been really interesting about Shopify in particular is this huge ecosystem of apps to enhance it. That’s incredibly powerful, but when you have all these apps interacting with each other, sometimes they interact in ways you can’t see.

We see this a lot. Someone installs an app, and it breaks another app. That can get really scary when it comes to the signal that’s powering all your paid advertising.

One of the things that really differentiates us is that we have strong anomaly detection in our application. We’ve put a lot of effort into understanding what’s happening, because we’ve come in and rescued a lot of companies—or heard horror stories—where a brand is gearing up for a big sale, they make a change, and then they’re not sending any signal into Meta for a week. They’re spending money on advertising, but because Meta isn’t getting signal back from the website about who’s visiting and who’s converting, it can’t optimize its spend.

That’s a big thing we do. We’ve built a strong reputation for being transparent, proactive, and making sure our brands can focus on their work and trust that the data is solid.

The other side of it is just a lot of hard, nerdy optimization work. In this space of improving signal in a “cookieless” world, there’s a lot of marketing speak and a lot of people claiming to do things nobody else can do. That’s not really how it works.

There are best practices for sending data to advertising platforms and techniques that evolve over time, but underneath it, it’s a lot of work to make sure every bit of identifying data that can help improve Facebook’s algorithm actually gets sent.

One thing high-end products like ours—and we’re not the only ones, companies like Elevar and Blotout are in the same category—do is implement what we call identity graphs. These are tools that, when we’re sending data to Meta, TikTok, Klaviyo, and others, let us stitch together personal identifiers for customers signing up for newsletters, logging in, joining loyalty programs, and so on.

We’re able to capture all that first-party data and send it to these platforms to help them better identify the people interacting with their ads. We all do it a bit differently, but at our core it’s similar: tireless work in our apps to make sure we’re not letting signal slip through, and that Meta and the other platforms are getting the best data possible.

Steve Hutt:
Yeah. When I think about server-side versus pixels, I know the difference because I’m in this industry. But some people listening might know the language and not the technical side.

A lot of merchants are probably still relying on browser-based pixel tracking. From my perspective, with ad blockers, iOS privacy changes, cookie restrictions, and so on, how much signal do you think is actually being lost right now? And how is server-side tracking fixing that signal loss?

Sean Larkin:
Yeah, that’s a great question. In short, if you’re just relying on pixels, the signal loss is catastrophic.

Steve Hutt:
Oh, wow.

Sean Larkin:
Back in the early days of Meta advertising—say 2010 to 2015—when Facebook ads really came into their own, you could place a pixel on your site. Facebook gave you a snippet of code, you put it on your site, and Facebook would use that to track you across every site you visited. That’s the classic third-party cookie, tracking you across many domains.

Since around 2016–2017, Apple came out with major iOS privacy updates and completely changed the game. Those third-party trackers that followed you across sites were essentially killed.

You could still do pixel tracking, but only on specific sites—your own site. You could track engagement on your site, but that pixel wasn’t tracking users across the internet anymore. That was just the first step.

Every year since then, cookies have been more and more restricted in what they can capture. JavaScript on your site is increasingly limited. You’ve got browsers like Brave with built-in ad blockers, more and more U.S. consumers using ad blockers, and Chrome blocking third-party cookies.

On top of that, newer iOS versions—23, 26, and so on—are stripping identifiers out of URLs. You send a Klaviyo email, and the campaign ID in that URL can get stripped out in mobile browsers depending on the settings.

So it’s getting really hard to rely on pixels alone. That’s where server-side tracking comes in.

There are a lot of different definitions of “server-side,” and the term has definitely been co-opted in marketing speak. But for folks who don’t know, we’re often talking about things like Meta’s Conversions API.

Practically, what that means is capturing events—some in the browser, some directly from Shopify via software like Fueled—combining them, and then sending those events server-to-server from our servers to the platforms’ servers.

We’re not unique in doing this, but the big benefit is that Fueled doesn’t get blocked as aggressively by ad blockers as a Facebook pixel. Browsers know what the Facebook pixel is and can easily block it. Fueled, Elevar, Blotout—we’re not targeted by ad blockers to the same extent.

And because we supplement with signal coming directly from Shopify, even if our pixel got blocked in the browser, we’re still receiving events from Shopify that we can send to Meta, TikTok, and others.

The key is that we’re not interrupted by network connectivity issues, someone’s internet cutting out, pages loading slowly, or ad blockers in general.

The other thing server-side tracking unlocks is the ability to enrich data with an identity graph like I mentioned earlier. In the browser, we help manage the Facebook pixel and improve the signal going into it by adding identifiers where possible, but we’re limited to about a seven-day window because of cookie deletion.

With server-side tracking, we can extend the attribution window for years and continue telling Facebook as much as possible about who these customers are. At the end of the day, what matters is performance improvement.

With our standard self-serve solutions, brands typically see a 20–30% improvement in ROAS on Meta because of those extended attribution windows and the extra customer data we’re able to send.

Steve Hutt:
I see. I’ve seen a lot of these hidden problems firsthand—clients coming to me at Shopify talking about misfires, duplicate events, missing conversions. In some cases, the disparity between platforms is more than 20%.

Those hidden issues silently hurt performance, and I think that’s a really interesting problem your solution—and a few others in the market—helps solve. When you have direct server-to-server communication, you finally have a source of truth, and you can feel confident the data is correct.

Then, when you’re spending on paid ads across different networks, you’ve got proper data to make decisions. Even when you’re trying to find net-new customers, you want to know you’re getting the right data because there are so many different campaign types.

Sometimes you’re running prospecting campaigns to find the next new customer who hasn’t purchased yet, and you want to enrich the data you already have and pipe that into separate campaigns. I know your platform can do that too.

Sean Larkin:
Yeah, for sure. It’s interesting to talk about when tools like Fueled become really valuable for a merchant.

If you’re just getting started on Shopify, you’ve made your first $1,000 or your first $10,000 month, tools like Fueled and our competitors aren’t necessarily what you need yet. The out-of-the-box Facebook integration for Shopify is solid. It’s well maintained. Sure, it has the occasional outage—like all software—but it’s fine.

Where tools like ours really come into play is when you start spending enough on ads that you’re seeing less return on each new dollar. There’s a natural saturation point in paid ad spend where, no matter how big you are, you eventually hit diminishing returns.

For many brands, that might be around $50,000 a month in ad spend, or $100,000 a month. It depends on your product, competition, and other factors.

Typically, most brands that come to us are doing around $8 million or more in GMV. They’re graduating out of small-business territory and starting to hire dedicated staff.

That’s where we come in. If you think of the saturation curve—a curve where performance improves, then starts to flatten and drop—our job is to shift that entire curve up. We provide better data so you don’t hit the crest of that curve as quickly, and you can keep scaling ad spend efficiently because you’re sending better and better data as you go.

I don’t envy merchants. On the side, I’m a goat farmer—I raise goats and make goat milk soap. I actually tried to do that professionally and then saw the first Dr. Squatch YouTube video six or seven years ago and thought, “Okay, I’m not going to compete with that.”

But that experience is what got me into Shopify. Merchants have it hard. They have to focus on creative, tactics, operations—so many things. The last thing they want is to read obscure documentation on how pixels are supposed to work.

Even with really big brands like Dr. Squatch, Caraway, Cuts Clothing, Built Basics, and Boats Overnight—these are sophisticated teams, but they can’t be experts in everything.

Most of our clients come to us because we pair high transparency with strong tech. We give them a demo, we handle the installation, and we do ongoing proactive monitoring and check-ins, usually once a month.

They simply trust that everything in that invisible layer is running correctly, and that’s the real value: improved ROAS with a high level of trust so they don’t have to worry about it.

Steve Hutt:
That’s amazing. I want to talk about the Opensend acquisition because I think it’s interesting how the two companies are coming together.

For merchants who might know Opensend—as I’ve mentioned them before and had them on the show—the platform helps brands and even media companies like mine identify anonymous website visitors.

Can you explain what it actually means to put these two pieces together? You’ve used the term “identity graph,” and now you’ve got a strong data pipeline. I think we’re going to be in a CDP conversation in a minute because there’s interesting stuff happening with data.

But first, help us understand the acquisition and what it means to bring Fueled and Opensend together as individual products that are now combined.

Sean Larkin:
Yeah, it’s incredibly exciting. I got to know Opensend’s founders about a year ago. We clicked quickly and started building projects together, combining our software. It became obvious there was something really cool there.

I wasn’t looking to get acquired, and I didn’t know much about the identification space. I knew what Opensend was, I knew Retention and some of the other players, but I wasn’t deep in it.

It’s an interesting space. It can be extremely powerful and build great customer experiences when done well, but if identification isn’t accurate and you’re identifying the wrong people, it can cause problems—like hurting your email sender reputation.

What became clear was that Opensend was approaching this in a very sustainable way. They’re incredibly transparent about how they do identification, where they source their data, and they have a strong chain of custody around opt-ins.

Once I saw that and the level of candor they had about how they solve anonymous visitor identification, any concerns melted away. I realized, “These guys are doing it right. Their data is extremely reliable.”

That opened the door to all sorts of cool things we could do together.

Historically, in the identification space, companies like Opensend and others focused mostly on getting net-new email addresses you could plug into your email platform. That was often the main use case.

What’s amazing about combining our two companies is that we can use Opensend’s identification tools to further boost identification inside your advertising platforms. That’s just one use case, but it’s a big one—and we’ve already built it.

I mentioned that Fueled improves ROAS on Meta by 20–30%. When we layer in Opensend’s identification, we can further identify anonymous visitors and send hashed email addresses and phone numbers to Meta to boost event match quality scores.

We’re seeing dramatic lifts. We’re taking strong scores—like a 6.9 in Facebook Events Manager—and pushing them to around 7.9 for upper-funnel events. If you haven’t been in Events Manager, that number might not mean much, but it’s basically a score of how good your signal quality is.

In plain language, we’re able to tell Meta who 30–40% more of those people are by providing email and phone identifiers. Meta can then say, “Oh, that’s Sean,” or “That’s Steve on his iPad,” and deliver remarketing across all of Steve’s devices.

When you combine strong identification with really robust data pipes, there’s a lot of powerful stuff you can do. Honestly, I get chills talking about it. Some of it I don’t want to share publicly yet.

Our Meta CAPI integration for customers using both tools already has that enrichment built in. Next week we’re launching a TikTok integration with Opensend enrichment, which will be huge. TikTok is a great emerging platform, but it’s not as big as Facebook, so better identification of anonymous visitors can have a major impact.

We’re already doing those things, and where we’re headed is asking, “How do we combine identification and Fueled’s pipes to build dynamic, personalized experiences across channels?”

One thing Opensend hasn’t talked about much yet is that, in addition to the identity graph of hundreds of millions of addresses and emails, we also have about 1,500 data points on 300 million U.S. consumers.

That’s the next level: appending that data and using machine learning on top of it to do two things with Fueled. One, improve targeting in ad platforms by sending custom events to Meta to help you find new customers. Two, enrich the events being passed so you can build more personalized experiences and smarter segmentation.

Steve Hutt:
That’s amazing.

Sean Larkin:
Every day, I’m just excited. I haven’t had a “job” in 18 years—I’ve run agencies and sold agencies—so this is the first time I’ve had a boss and coworkers. I love it because we’re building really cool stuff.

We encourage folks to stay in touch with us because this is going to be a really big year in terms of new functionality.

Steve Hutt:
I totally agree. The union of the two companies makes a lot of sense, and I’m excited to see where it goes.

It sounds like there’s going to be another podcast episode after the summer—maybe early fall—once some of this next wave of functionality is live. I like to be first out with this kind of story, so we can chat under embargo and then release it when the time is right.

Sean Larkin:
Yeah, you can have it first.

Steve Hutt:
Love it.

Let’s talk about larger brands that might be listening—more enterprise-level. Some of them already have their own CDP. Klaviyo has one now, Yotpo is working on theirs, and there are a lot of platforms calling themselves enterprise customer data platforms that aim to be the single source of truth for the customer record.

They might also be using Segment or mParticle.

Can you talk about that? There are six-figure contracts and engineering teams attached to those enterprise CDPs. Do you think Fueled is an interesting solution for brands that are mid-market to enterprise—and with the union of Opensend and Fueled—is this a direct replacement for a CDP?

Sean Larkin:
Yeah. Yes.

We grew up, so to speak, as a Segment connector. Fueled started as a small cottage business because I was running an agency and we had a BigCommerce client who wanted to integrate BigCommerce with Segment using server-side tracking—and that integration didn’t exist.

So we built it. Then we built a Yotpo Reviews integration for Segment, then a Loop Returns integration. At that point we had all these sources pushing events into Segment, and Segment was partnering with us a lot. We thought that’s where we’d stop.

Enterprise CDPs like Segment are incredible, but they have two main challenges.

First, when you’re talking about an identity graph, it’s hard to store identifiers and build them directly into the data flow. This isn’t a Segment or mParticle problem specifically—it’s just the nature of big enterprise CDPs. It’s hard for them to do the kind of enrichment we do.

Second, they can be difficult to get right. They’re big platforms. There’s a lot of point-and-click configuration. There are only a handful of agencies in the world that really know how to implement them properly for ecommerce.

For B2B or other industries, they’re fantastic. But for ecommerce, there’s a right way to integrate with Facebook, and it’s not always obvious.

Where we differentiate is that we have a strong point of view on how to do this right for ecommerce, and we just do it out of the box. I’ve worked on some really cool Segment implementations that do things nothing else can, but they require weeks or months of development.

With us, you come in and it works out of the box. You’re up and running the next day—even with headless sites.

It’s interesting to see Shopify Hydrogen and where headless is going. I come from the Drupal world, and we’ve been talking about headless for 20 years. It’s always “the next big thing,” but we’ll see. It’s a very specific solution for specific needs.

Steve Hutt:
Yeah, it dropped off a bit. I remember people using headless with Shopify or Webflow, but the challenge was refactoring both front ends—Shopify on the backend and something like Contentful on the front—and making them work together.

There were delays and issues, and a lot of brands moved back to Shopify’s native stack. There are brands doing hundreds of millions on Shopify’s core platform without going headless. Everyone has their own desired customer experience, of course.

Sean Larkin:
Oh yeah, and Hydrogen works great for certain things.

We’ve built a strong integration for those use cases as well. We have an incredible Pack Digital integration out of the box. In about four hours, you can have the best possible signal in a headless setup for Meta, TikTok, and other platforms.

We’ve focused on those use cases for brands like Dr. Squatch and Caraway. Those aren’t everyday use cases for most of the Shopify ecosystem, but for the brands that are considering enterprise CDPs, we provide a lot of value.

We also have a Segment connector and a RudderStack connector—both enterprise CDPs. So if someone needs to use an enterprise CDP, and there are good reasons to do that, you can turn us on and we’ll ensure those platforms are getting the best possible signal.

We play in all these spaces.

During COVID, there was a big push to get enterprise tools in place. Now, with tariffs and cost pressures, people are trying to be more frugal. That’s where we add a lot of value—we’re priced for Shopify and very quick to turn on and show results.

Steve Hutt:
Yeah, I see that. The pricing is amazing.

So what does implementation and timeline actually look like? A brand installs Fueled today—what does onboarding look like? What happens in the first week or first 30 days? What do brands typically see once they install it? Is it set-it-and-forget-it? Do campaigns just become more efficient?

Sean Larkin:
Yes. Both.

We have a self-service version of our software listed in the Shopify App Store. It’s easy to install.

That said, a lot of the customers we work with reach out to us directly. We offer annual pricing, and when we do annual deals, we still bill through the Shopify App Store to keep it simple. For those annual plans, we handle onboarding for you.

For brands doing $10 million or more, that’s very attractive. Fueled is intuitive enough that I can install it on a Shopify site in about five minutes, but people want trust. They want to know they’re in good hands, that the software works, and that they don’t need to think about it.

For annual customers, we’ll typically do the onboarding. We get access to their ad platforms, their Shopify store, and their ecommerce email platforms. Then we schedule the install. It takes about 10 minutes, and we add a layer of monitoring and extra testing to make sure everything is perfect.

We also work in Slack with our customers. If they have a big sale coming up, they’ll ping us and say, “Hey, we’ve got a sale next week. Can you take a look at our setup and make sure we’re good?”

That kind of proactive support is unique. Of course we’ll log into Meta and double check everything. For Hydrogen and other custom enterprise environments, that’s especially valuable because there’s so much complexity. Being able to schedule a review anytime is a big deal.

Steve Hutt:
Yeah, that’s amazing.

Sean Larkin:
I come from the agency world, so being high-touch, transparent, and clear about how our software works is really important to me. It’s a big reason I gravitated toward Opensend—they share those values.

Steve Hutt:
For those listening who are thinking, “Maybe I need to look at my data infrastructure and figure out next steps,” what should they do?

You’ve got individual products, but they now work together. On the Opensend side, I know there’s an offer for my listeners—50% off your first month. I’ll put that link in the show notes. If you’re strictly looking for identity resolution to enrich your Klaviyo account with anonymous website visitors and run a “thanks for visiting” sequence, Opensend can do that.

Now with Fueled and Opensend combined, what do you see as the best next step? How should people reach out if they want to understand more about the two together? We’ve got the 50% off offer right now. I’m not sure what Fueled is offering on its side, but—

Sean Larkin:
Oh yeah, the same thing for sure.

Steve Hutt:
Okay, cool. So they just go to the website directly?

Sean Larkin:
Yeah. We finalized the acquisition literally a week ago, so you’re early. I still have both my Fueled and Opensend email open every day.

If you want to get in touch, you can go to opensend.com. We have a dedicated landing page for Fueled at opensend.com/fueled—like “I fueled my car.”

You can also go to fueled.io. Either site is a great starting point, and there are clear “book a call” links.

We’ve got five or six people from Opensend and Fueled headed to Shoptalk in two weeks. We’ll be drinking lots of coffee at Hazel’s, so folks should look us up. Ping us on LinkedIn or wherever—you’ll find us.

Steve Hutt:
Beautiful. I’ll update the show notes with all of that.

This has been a really impactful show. I’ve learned a lot. I had a reasonable understanding of cookies and pixel tracking versus server-side, but you’ve really closed the loop on the true opportunity—what’s available now and how you’re future-proofing what you’re building, especially around personalization.

It’s a great time to be alive. There are a lot of AI-driven enrichment tools helping on the data and code side. We talked offline before recording—this is a blue ocean of opportunity. There are plenty of competitors trying similar things, but I’d say most people are not using server-side tracking at all.

That’s the point I want to drive home: if you’re not using server-side tracking and you’re spending $100K a month on paid ads, there’s definitely some wasteful spending happening.

Sean Larkin:
Oh, for sure. Absolutely.

Steve Hutt:
As we wrap up, let people know what to mention when they reach out.

Sean Larkin:
Yeah—tell us you listened to this podcast. Tell us Steve sent you. We’ll take care of you.

We’re big fans of this podcast, and we want to support the community you’re building.

Steve Hutt:
Love it. Thank you.

It’s great to meet you, Sean. Thanks so much for recording today, and congrats on the acquisition and this next iteration of your career. It’s really cool. And welcome to the Shopify community.

Sean Larkin:
Thank you. I appreciate it.

Steve Hutt:
All right, have a great afternoon.

Sean Larkin:
You too.

Steve Hutt:
Take care.

Well, that’s it for today’s episode. I’d like to thank you personally for being a loyal listener of eCommerce Fastlane. It’s my hope that this podcast is giving you a ton of value through growth strategies, tactics, and exclusive insider tips on the best Shopify apps and marketing platforms—all with my goal of helping you build, manage, grow, and scale a successful, thriving company powered by Shopify.

Thanks for investing some time today and listening to the show. I’m so proud and excited that you have a growth mindset and are a constant learner. I truly appreciate you and your entrepreneurial journey.

Enjoy the rest of the week, and keep thriving with Shopify.