[00:00:00] ​Intro

[00:00:26] Phil: What's up, everyone? Today, we have the pleasure of sitting down with Steven Aldrich, co-CEO and co-founder at Ragnarok NYC.

[00:00:34] About Steven

[00:00:34] Phil: Steven's first job after business school was as a customs broker in Colombia before his visa ran out, and he had to return to the U.S. He started his marketing career as a marketing and comms associate at a market research firm, where he discovered the wonders of HTML, email development, and Adobe Dreamweaver. Back then, while still in his full-time job, he took on email and CRM roles across various industries. Steven and his co-founder Spencer launched Ragnarok first as a side hustle, moonlighting as martech consultants. But in 2017, both co-founders took the leap and went all in on the agency.

[00:01:13] Today, Ragnarok is a 50+ person full-service martech agency that's worked with brands like Zapier, Dropbox, Asana, Adobe, and many more. Steven, super pumped to chat with you today. Thanks for being here.

[00:01:24] Steven: Oh, Phil, great to be on here. As I mentioned before, I'm a huge fan of the show. Been a listener for a little over a year now. It’s kind of weird to be on the other side, honestly, because now when I listen, I’ll be like, "Oh, now I know what Phil looks like!" That’ll be interesting. You mentioned your "radio face," but now I can see you while you're talking in my ear while I’m walking my dogs or doing whatever else.

[00:01:51] Phil: It's always fun hearing from someone who’s listened to the show before coming on. I’ve been on the other end a few times, and I love to listen to past episodes to get a feel for the types of questions. Sometimes I get lucky and catch the fast questions that come at the end of the show and prep ahead.

[00:03:56] Steven: Oh yeah, I always have the worst "fun facts" because I should have one ready, but, you know, I never do. But yeah, I know you’ll ask the one at the end, so we’ll save that bit.

[00:04:16] Phil: Nice, right on. So, the first question is a fun one. I feel like you’re the perfect person to ask this. Like I said in the intro, Ragnarok has been around for years, and I think you’re one of the first martech-exclusive agencies.

[00:04:27] The Evolution of Martech and the Impact of Consolidation

[00:04:27] Phil: Marketing agencies have existed since the dawn of the internet and even before, but branding agencies and web agencies have been more common than martech-exclusive ones. But you guys have been around since 2012—basically when I started in marketing. Back then, marketing automation was just taking off, and you could still see all the logos on Scott Brinker's martech landscape without a magnifying glass, unlike today. Knowing how much and how fast the space has changed in the past decade, what do you think the next decade holds for us? Is martech going to keep exploding as a category? What does your crystal ball say, Steven?

[00:05:18] Steven: Yeah, you know, it's funny. I get these visionary questions a lot. The first thing I usually say is, "Well, what would Scott Brinker say?" or "What would someone who researches this every day think?"—before forming my own opinion. I have lots of opinions, but it’s helpful to see what others say. I’ve probably used around 150 martech tools myself, but there are tens of thousands out there today.

[00:05:46] Where I see the category evolving is leaning towards best-in-breed stacks, particularly in a consolidation phase right now. We saw the big acquisitions, like Twilio buying Segment or Salesforce pushing its CDP. ContentSquare recently acquired Heap, for instance. We're also seeing some slowdowns after the tech meltdown of 2022-2023, but there’s still steady activity.

[00:07:00] The next few years will likely see this leveling out, with newer AI-based martech startups maturing and either getting acquired or standing out as winners. The Gen AI wave is coming in hot, but once that becomes table stakes, it’ll shift to how Gen AI gets enabled and optimized. Early players like Jasper or Frees.io focus on making AI brand-focused, and you’ll eventually see marketers leveraging tools like Iterable’s Gen AI journey builder more easily.

[00:09:00] In the next 10 years, I believe we’ll shift from marketers orchestrating entire journeys to more rule-setting where the system runs itself. We already see basic automation for things like abandoned cart emails, but as AI integrates deeper into martech, the focus will be on the marketer defining strategy while the systems handle the execution.

[00:09:20] Phil: Yeah, it’ll be fun to see how that evolves. I’ve had a lot of conversations around AI and personalization. There’s this spectrum, right? Some people want full control over every message, while others just want the machine to take it all off their plate.

[00:09:51] Blending Automation with Human Spark for Smarter Martech Strategies

[00:09:51] Phil: I’ve had a lot of conversations with folks around that. There's this wide spectrum—three different camps, really. On one side, you’ve got people who believe machines will never know their audience like they do and prefer to write and send all their own messages. On the other extreme, you have people who are like, "Just let the machine handle everything while I go grab a cup of coffee."

[00:10:28] But the more interesting group is the one in the middle. These people want to leverage AI for automating optimization but still want a human in the loop for quality assurance. Especially for companies in industries like fintech or healthtech—like my most recent role—where there are serious consequences if the wrong message gets sent at the wrong time. You can’t just let the machine run free in HIPAA or financial regulatory environments. So, yeah, it’ll be interesting to see how this plays out.

[00:11:22] Steven: Yeah, I think there’s always a baseline of what everyone does, right? Like, abandoned cart programs are now so standard. Everyone runs the same flow—an email within two hours or so, usually powered by something like Klaviyo. You don’t even have to think about it anymore; it’s table stakes.

[00:12:06] But, when it comes to industries like fintech or healthcare, you’re right—machines can’t fully handle it. There’s always going to be a legal or compliance review needed at the end. And while the operational efficiency AI brings is undeniable, certain things—like knowing the exact right message for a highly regulated audience—still need that human oversight. It's all about identifying the pain points and understanding where automation can truly make a difference without sacrificing quality or compliance.

[00:13:34] Phil: Exactly. And this concept of incremental testing—automating small, continuous optimizations—is a huge opportunity for AI. But I think a lot of teams still want someone at the helm to make sure everything aligns with brand and compliance guidelines. Like you said, especially in finance and healthcare, the stakes are just too high for mistakes.

[00:13:57] Steven: Yeah, definitely. You know, even in less regulated industries, there's a balance between operationalizing things and making sure you're still crafting an authentic experience. And while automation handles repetitive tasks, like sending messages, it doesn't replace the thoughtfulness behind them. That's where marketers still play a critical role.

[00:14:30] Phil: Right. And as AI becomes more table stakes, marketers will probably focus less on the minutiae and more on higher-level strategy—like crafting experiences that really resonate.

[00:16:22] The Value of Data Science in Martech Optimization

[00:16:22] Phil: You touched on something interesting earlier about how large teams and resources can drive better automation strategies. I’ve seen that difference firsthand between startups and larger companies. When I was at WordPress.com, we had dedicated data scientists on the marketing team, which allowed us to build our own internal CDP and create some crazy powerful propensity models.

[00:17:00] We could track any event and forecast which users were most likely to convert, creating uplift models that allowed us to be super efficient with our marketing spend. But, like you said, most startups just don’t have those resources. Do you think there’s a future where more of those enterprise-grade capabilities are democratized for smaller companies?

[00:18:00] Steven: Yeah, that’s a great question. And I agree, what you had at WordPress.com is super powerful. Having a dedicated team of data scientists who can build custom models for marketing is a luxury that very few companies have.

[00:18:16] For startups, it’s definitely harder to replicate that. But there are tools coming out that are trying to make those capabilities more accessible—tools that decentralize what larger enterprise teams do. And we’re seeing some movement there. But it’s still a big challenge, especially when it comes to smaller datasets. It’s tough to build accurate models without enough data to train on, so that’s where AI’s potential comes in. As AI gets better at processing smaller data sets, we could see more startups using these advanced capabilities.

[00:19:08] Phil: Yeah, it’s a huge opportunity. Some larger companies like Salesforce are already automating email batch campaigns based on AI-driven messaging decisions, with no human intervention. Startups are far from that level, but it’s interesting to think about how AI might help close the gap in the future.

[00:19:08] Steven: Yeah, exactly. I’ve been part of organizations that built those types of AI-driven email journeys where the system makes the messaging decisions. The marketer’s role shifts to more of an asset manager, feeding content into the system. It’s powerful, but like you said, it requires a lot of data. That’s the limiting factor for smaller companies.

[00:20:00] It’s a channel of scale. AI can help with timing, and you’ll see studies from marketing automation tools that say, "We increased conversion by X percent by optimizing send time." That’s where AI shines, but you still need humans to craft the overarching strategy and make sure the machine is aligned with your brand voice and goals.

[00:21:00] Phil: Exactly. The human element still plays a huge role, especially when it comes to things like brand tone and personalization. But as AI evolves, it’ll be fascinating to see how much it can take on and where marketers will continue to be essential.

[00:27:50] A Nuanced Approach to Evaluating Martech Vendors

[00:27:50] Phil: You mentioned a bunch of tools earlier. One of the things I admire about Ragnarok is that you guys are platform-agnostic. You don’t pigeonhole yourselves into one particular stack, but you do have a shortlist of preferred vendors. How do you go about evaluating new tools when they come up? How do you decide which vendors to add to your list?

[00:28:24] Steven: Yeah, we do get a lot of flak from partners about this—"Why do you support our competitors?"—but we’re exposed to so many tools that it’s impossible to have a strict bias anymore. Once you’ve worked with six or seven or more platforms, you start seeing the strengths and weaknesses of each, and you learn to evaluate them more holistically.

[00:28:55] A good example is Insider, one of our newer partners. They’re based out of Europe and built this really interesting platform that combines CDP and marketing automation with out-of-the-box recipes for sophisticated e-commerce use cases. When we evaluate tools like this, we look at a few core things: Do they have APIs? Can I get data in and out easily? Do they support event-based data? How modern is their infrastructure in terms of bandwidth and throughput? These are our table stakes.

[00:29:55] After that, we look at their unique value proposition. What are they building that’s different? Is there a specific market or industry they cater to? And what’s their long-term vision? That’s really important because it informs whether they’ll be around in a few years or whether their product will stagnate.

[00:31:32] Tools like Braze have carved out a niche by focusing on the marketing experience and integrating well with CDPs and reverse ETL tools. I love their focus, but I also appreciate when a tool like Segment focuses on data enrichment and orchestration without stepping on Braze’s toes. It’s all about where these vendors fit into the broader ecosystem and how they play with other tools.

[00:33:00] Phil: I love that you’re not just evaluating tools for your clients but that these martech vendors actually involve you in their roadmap decisions. You’re in this unique spot where you’re advising clients on the tools they need, but also giving feedback to vendors on what’s working and where they can improve. How much advisory do you do for the vendors themselves?

[00:33:00] Phil: I love that you’re not just evaluating tools for your clients but that these martech vendors actually involve you in their roadmap decisions. You’re in this unique spot where you’re advising clients on the tools they need but also giving feedback to vendors on what’s working and where they can improve. How much advisory do you do for the vendors themselves?

[00:33:15] Steven: Yeah, it's definitely a unique spot to be in, and we do a fair amount of advisory with the vendors themselves. We give feedback on roadmaps, new feature ideas, and overall product direction. I’m always upfront with the product folks, though—I tell them, “Brace yourself, I’m opinionated.” I’ll tell them exactly what I think because, at the end of the day, our clients depend on us to recommend tools that can evolve with their needs.

[00:33:45] A good example of this is Klaviyo. When they were building out their CDP and mobile capabilities, they reached out to us for input. They wanted to know what features were table stakes versus what would truly differentiate them in the market. That’s a great position to be in because we have insight into both sides—the client pain points and the tech vendor’s roadmap.

[00:34:15] We’ve also worked closely with Braze on their certification exams, and I’ve been involved with Iterable in advising on how they think about mobile and data ingestion. The partnership with Hightouch, for example, was a big step in the right direction. These collaborations can really help shape the tools in ways that directly benefit our clients.

[00:34:55] One of the most exciting aspects of this is when we help vendors forge partnerships with other companies. Sometimes, Martech vendors don’t realize how much synergy they have with another player until we connect them. For example, there are times when two companies might view each other as competitors, but after some conversations, they realize they could complement each other’s strengths. That’s where we come in, facilitating those partnerships and making those connections that can bring mutual value to both sides.

[00:35:45] Phil: That’s super cool. And I imagine being in that position gives you a unique lens on where the market is going and what kinds of features these tools will need in the future.

[00:36:00] Steven: Absolutely. It’s like being a translator between what the market needs and what the tools are building. We see what clients are asking for, and we can tell the vendors, “Look, you need to focus here because this is the next big thing.” It’s a win-win situation because it keeps the vendors relevant and helps our clients get better tools that fit their evolving needs.

[00:36:45] Phil: That makes a ton of sense. It’s like you’re aligning both sides—what clients need today and what vendors are building for tomorrow.

[00:37:00] Uncover the Martech You Actually Need with a Bottom-Up Approach

[00:37:00] Phil: Switching gears a bit—you mentioned earlier about clients coming to you, saying, "We need a tool, but we don’t know what to buy." I think this is a pretty common challenge, especially when teams don’t fully understand what they actually need versus what they think they need. How do you help clients figure out what Martech they really need to solve their business use cases?

[00:37:25] Steven: Yeah, we call this our “bottoms-up Martech analysis.” It's something we've honed over the years, and the key is to ground everything in actual business use cases. You don’t want to start with features, because you’ll end up buying tools that sound cool but don’t actually solve your problems.

[00:37:45] Instead, we start by identifying what the business is trying to accomplish. For most companies, that breaks down into 50 to 80 use cases. These are things like abandoned cart programs, re-engagement campaigns, or personalized content for high-value segments. Once we’ve identified those use cases, we can build an evaluation matrix that maps those to potential tools.

[00:38:15] From there, we work with the vendors. We give them specific scenarios and ask, “How would your platform handle this?” That’s a critical part of the process because it forces the vendors to focus on real-world problems instead of pitching us features we might never use. It also makes it easier for our clients to evaluate vendors based on their ability to solve real pain points.

[00:38:45] We also use a scoring model to compare vendors on how well they address each use case. This helps us get into the nuances that might not be obvious at first glance. For example, two tools might seem similar in their email sending capabilities, but one might have a slight edge in SMS or push notifications, depending on the client’s needs.

[00:39:10] Phil: I love that approach. It’s so easy to get caught up in features that sound cool but don’t actually move the needle. Grounding everything in business use cases forces the conversation to be more practical and focused.

[00:39:25] Steven: Exactly. We always remind clients that you’re not buying a tool for today; you’re buying it for the next three, five, even ten years. So it’s important to evaluate how well the tool can grow with you, whether that’s through new features or through their vision for the future.

[00:42:28] Uncovering New Use Cases by Aligning Martech with Real-World Needs

[00:42:28] Phil: That’s such a solid approach, and I imagine working closely with both the marketing and data teams helps uncover new use cases that the marketing team might not have thought of. When you bring those insights to the table, it must change the way clients think about their Martech stack, right?

[00:42:55] Steven: Oh, absolutely. One of the most rewarding parts of the process is when we can surface new use cases that weren’t initially on the client’s radar. A lot of times, marketing teams will come in with a narrow view of what they think they need. But once we involve the data team and dig into their existing infrastructure, we start uncovering capabilities they didn’t even know they had.

[00:43:30] That’s when the conversation gets really exciting. We can go to the marketing team and say, “Hey, did you know you already have the data infrastructure to power XYZ use case?” And then we can look at tools that will help them activate that data in a way that drives real value.

[00:43:55] Phil: That’s huge. I love that you’re helping teams unlock new potential by bringing those insights to the surface. It must make the Martech shopping process way more strategic.

[00:44:10] Steven: Yeah, it definitely does. It changes the conversation from, “What tool can solve our immediate problem?” to, “What tool will enable us to do more in the future?” That’s where the real value comes from. You’re not just buying a tool to check off a box—you’re investing in something that will grow with your business.

[00:47:00] What the Best Martech Implementations Prioritize

[00:47:00] Phil: So, you’ve helped the client select a tool and start implementing it. What’s your approach to ensuring they get the most value out of their Martech investment?

[00:47:15] Steven: Once a client has chosen a tool, our priority is to help them realize value as quickly as possible. The best implementations are the ones where you’re in market with your first use case within 90 days. Sometimes even faster, depending on the complexity of the business.

[00:47:35] The key is to stay focused. You don’t want to bite off more than you can chew. You need to get something out the door that drives value quickly, and then build on that momentum. Too many teams get bogged down trying to perfect everything before launch, and they lose sight of the goal—which is to drive value as soon as possible.

[00:48:00] Our rule of thumb is that the investment should pay itself back within two years. The first year might just break even because of the resources you’re putting in, but by year two, you should see significant ROI. By that point, you’ve deployed enough use cases to have real business impact.

[00:48:30] Phil: Makes total sense. Start with something simple, prove value, and then build from there.

[00:48:35] Steven: Exactly. And when you’re measuring ROI, it’s not just about the tool—it’s the total cost of ownership. You have to factor in the time and effort of your team, as well as any external resources you’re using. That’s why it’s so important to launch quickly and start seeing returns early.

[00:54:01] Measuring Marketing and Martech Impact Using the BEATS Model

[00:54:01] Phil: I know we touched on ROI earlier. I’d love to dive deeper into how you measure the impact of marketing and Martech. You mentioned multi-touch attribution and incrementality earlier. How do you approach measuring marketing’s impact and balancing things like brand, time lag, and causality?

[00:54:55] Steven: Yeah, I’m a big fan of the BEATS model. It’s a framework that helps ground marketing measurement in business outcomes. The problem with a lot of marketing metrics is that they get too granular and lose sight of the bigger picture. The BEATS model starts with the business—what’s the business outcome we’re trying to drive?—and works down from there.

[00:55:30] You start with the business outcome, then you look at experiments you’re running, then you layer on attribution models. The key is to always keep the business outcome in mind. It’s easy to get lost in the weeds with multi-touch attribution or incrementality models, but if you’re not driving real business results, none of that matters.

[00:56:20] Phil: That’s such a good point. Sometimes it feels like attribution models are just slicing the data in a way that looks good but doesn’t actually move the needle.

[00:56:35] Steven: Exactly. You need to understand the levers that are actually driving business outcomes. And while multi-touch attribution is great for understanding touchpoints, you can’t lose sight of the fact that the business needs to grow. That’s why I love the BEATS model—it keeps you focused on the big picture and helps you identify the levers that actually matter.

[00:57:00] Phil: That’s a great perspective. I’m definitely going to look deeper into the BEATS model because it sounds like it cuts through a lot of the noise.

[00:57:10] Steven: Yeah, it’s a helpful framework, especially when you’re working with executive teams that want to see business impact first and foremost. Attribution is important, but it’s just one piece of the puzzle.

[00:59:00] Finding Balance Between Career, Family, and Happiness

[00:59:00] Phil: We’re running up against time here, but I wanted to wrap up with our signature question. You’re juggling a ton of things—co-CEO of an agency, podcast co-host, father, and you’ve got all these other hobbies like amateur carpentry and ballet. How do you manage it all and stay happy?

[01:00:00] Steven: That’s a great question, and I think a lot of it comes from my upbringing. I grew up in a blue-collar family, and I was the first person in my family to graduate from college. That background gave me a strong work ethic, but it also helped me appreciate the simple things.

[01:01:00] For me, happiness comes from the life I’ve built. I’ve always wanted to be a dad, and now I am. I have a house, a family, and a career that I love. It’s not always easy—there’s a lot of chaos with two dogs, two cats, and a young kid running around—but at the end of the day, I can look at my life and say, “I built this.” That’s incredibly fulfilling for me.

[01:02:00] It’s about being in control of my own happiness. I don’t measure my success by comparing myself to others. I measure it by the life I’ve built and the things that matter to me. That’s what keeps me grounded and happy.

[01:02:44] Phil: That’s a great perspective, and I think a lot of people will resonate with that. Thanks so much for sharing your time and insights today, Steven. I feel like we could’ve gone down seven or eight different rabbit holes, so maybe we’ll have to do a round two sometime!

[01:03:00] Steven: Absolutely. It was a blast, Phil. Thanks for having me on, and I look forward to catching future episodes of the show.

[01:03:10] Phil: Awesome. Cheers!

[01:03:15] Steven: Cheers!