Speed isn't insight: Kevin Steba on what email analytics is really for
You’ve probably already faced this problem: Your analytics team now has a new tool, but reporting hasn't become any faster. Or it's become faster to get answers to your questions, but it’s still unclear whether the decisions based on this data are any better. At some point, you may also have wondered where AI's limits are in analytics and whether you can trust the information it provides.
We spoke with Kevin Steba, the cofounder and CEO of email and CRM analytics platform SEINō, and asked for his perspective on using data in decision-making.
Key takeaways
- Email is one of the few channels where creative decisions tie almost immediately to business results. You can test an idea and see the impact the same day.
- Treat open and click rates as diagnostic signals for spotting trends or technical issues, not as measures of success. The KPIs that matter are the ones tied to business outcomes: transactions, revenue, revenue per email, and customer lifetime value.
- Data earns its keep when it challenges what a team believes, not when it confirms it.
- AI won't replace a marketer's judgment, and its value depends entirely on the data behind it. Grounded in trusted data, it surfaces patterns a specialist might miss.
Expert
Kevin Steba is an email analytics specialist from the Netherlands and the cofounder and CEO of SEINō, an email and CRM analytics platform that helps marketing teams make better decisions through data. With over 20 years of experience in digital marketing, CRM, and analytics, he has worked on both the agency and client sides.
Kevin is a frequent speaker at industry events, where he shares practical insights into email marketing and analytics. He is passionate about helping CRM teams understand, trust, and act on their data.
How data and code pulled in a front-end developer and kept him in email
Stripo: You've built companies from the age of 17, worked in development, CRM, campaign production, and now analytics. Looking back, what pulled you specifically into email marketing, and what still keeps you in it after all these years on the operator side?
Kevin: It actually started after I sold my digital agency in 2017. I joined the CRM team at Schiphol Airport as a freelance front-end developer to rebuild all their email templates and lifecycle campaigns.
Coding email templates was something most front-end developers hated back then (and maybe still do!), but I became fascinated by how data, design, and personalized content came together in a single channel. Email sits right at the intersection of technology, marketing, and customer experience.
What has kept me in email is that it's one of the few channels where creativity and measurable business impact are so closely connected. You can launch something in the morning and know by the afternoon whether your assumptions were right. At its core, email is simply about having personal conversations at scale. Whether that's through email or another channel in the future, helping businesses communicate more personally is what still excites me today.
When the signal breaks: Rethinking what email KPIs should measure
Stripo: When a metric like open rate breaks, simply redefining it is a tactic without a long-term strategy. That's a point you've made directly in your LinkedIn post. So in 2026, what does a genuinely durable KPI framework look like once the signal email marketers leaned on for two decades stopped being reliable?
Kevin: The most durable KPI frameworks separate diagnostic metrics from business outcomes. Open rate and click rate are still useful for spotting trends or identifying technical issues, but they should no longer be treated as primary success metrics.
The KPIs that really matter are the ones directly linked to business objectives: transactions, revenue, revenue per email sent, customer lifetime value, or, for B2B organizations, meaningful conversions, such as demo requests or contact forms.
Kevin Steba,
Every business is different, but the principle is the same. Start with the business outcome you're trying to influence and work backward. Engagement metrics help explain why something happened, while business metrics tell you whether your marketing actually created value.
Stripo: "Data-driven" has quietly become something teams hide behind: sometimes to launder a gut call after the fact, sometimes to stall in analysis paralysis. In your experience, where do you actually see data change a marketer's mind, versus just confirm what they'd already decided to do?
Kevin: I've seen teams blame poor campaign performance on creative when the real issue was audience fatigue, send frequency, or a list that had gone stale.
The biggest mindset shifts happen when data disprove something people were convinced was true.
Kevin Steba,
That's where AI earns its place, too — not by replacing the marketer's judgment, but by flagging the pattern nobody thought to check, like a specific segment that's been ignored for three sends in a row.
Stripo: Building a real KPI framework is different from picking a few metrics to report. What's the biggest mistake experienced email marketers make when they put together reporting for stakeholders?
Kevin: One of the biggest mistakes is reporting on outcomes without explaining what actually influenced them.
A marketer doesn't control every factor that affects revenue. Pricing, product availability, and inventory often sit with other departments. What marketers do control includes subscriber quality, audience growth, campaign frequency, timing, messaging, and creative assets.
Reporting becomes much more valuable when it separates the metrics you own from the metrics you observe. Only then can you have meaningful conversations about what to improve, instead of simply reporting numbers.
Here's how Kevin breaks that down:
|
Metrics you own |
Metrics you observe |
|
subscriber quality |
pricing |
Another common challenge is a lack of context. Looking at a single campaign in isolation rarely explains why something happened. To understand performance, you need to zoom out and look at longer-term trends, audience behavior, and historical benchmarks.
Stripo: A team can cut reporting time dramatically and still walk into the same meeting with the same dashboards that nobody acts on. So is "faster reporting" actually the bottleneck for most teams? And when reporting goes from hours to minutes, what do you actually see teams do with the time they free up?
Kevin: The biggest difference isn't that marketers save a few hours every Monday morning. It's that they stop treating a question as a week-long project. If someone in a meeting asks, "Why did the Tuesday send underperform?" the old answer was, "Let me get back to you." Now they open the segment live in the meeting, and answer it on the spot.
Saving time on reporting isn't the goal; making better decisions is. That changes what teams spend their time on: less time collecting and formatting data, more time testing new segments, rewriting subject lines that aren't working, or catching a deliverability issue before it becomes a quarter-long problem.
Kevin Steba,
Where AI earns its place in analytics
Stripo: You ran AI cocreation sessions with your own customers to shape SEINō's AI features, so you've seen up close what people actually want versus what sounds good in a pitch. The category is flooded with "AI for email" right now. In analytics specifically, where's the line between AI that sharpens campaign analysis and AI theater that just adds noise? Is there something AI can do for analysis that a skilled human analyst can't?
Kevin: Large language models aren't great analysts by themselves; they're great communicators.
The quality of the output completely depends on the quality of the data, the business context, and the analytical framework behind it. AI becomes genuinely valuable when it's grounded in trusted datasets, historical performance, mathematical models, and domain-specific knowledge.
What AI does exceptionally well is process enormous amounts of information, identify patterns, and explain complex findings in natural language. A skilled analyst can certainly reach similar conclusions, but AI can do it much faster and explore far more combinations than a human could realistically.
I don't see AI replacing analysts. I see it making every marketer capable of asking better questions and every analyst significantly more productive.
The value that remains when email building gets commoditized
Stripo: Email development is its own special pain: rendering quirks and dark mode. As someone who came up writing front-end code, do you think hand-coding email is on its way out now that AI codegen and modular component systems are maturing fast? And if the build gets commoditized, does the real competitive edge move entirely to data and strategy?
Kevin: Yes, hand-coding email as most people know it today is going away. AI and modular component systems are already good enough to handle the rendering quirks and dark-mode edge cases that used to eat up a developer's whole afternoon.
But understanding how things work will always be valuable, even if you no longer write every line yourself.
I'm a big believer in automating the work that gives you the most friction. The catch is that automation only creates value if you understand what it's doing. If you can't validate AI-generated code or explain why something works, you've just moved the risk somewhere you can't see it.
Kevin Steba,
Note from Stripo
Modular email design means building emails from blocks you already have: headers, footers, buttons, product cards, banners, and more. Instead of creating these elements from scratch every time, you build a library of modules and reuse them whenever you need them.
It also makes sense to have different modules ready for different campaign types, whether you’re working on a promotion or a triggered campaign. When it’s time to send a new email, you just pick the blocks you need and put them together.
In Stripo, you can also synchronize modules. Update a module once, and the changes will appear in every template that uses it. You can also "lock" a module's design to prevent unwanted edits, and add AMP and other interactive elements to your modules. None of this requires technical skills.
Stripo members cut email production time by more than 3.7 times, while keeping branding consistent across every email they send.
So yes, the build gets commoditized. What doesn't get commoditized is knowing your audience well enough to know what to build in the first place. That's where the edge moves: understanding customers and making the strategic call, not typing faster.
Stripo: Are modern email marketers becoming too specialized, or is cross-functional knowledge still essential for building high-performing programs?
Kevin: Specialization makes you efficient, but curiosity makes you valuable.
The best CRM professionals I've worked with understand a little bit of everything. They know enough HTML to understand email development, enough analytics to interpret data correctly, enough UX to improve customer journeys, and enough business strategy to understand why campaigns matter in the first place.
You don't need to become an expert in every discipline, but having cross-functional knowledge helps you connect the dots others miss, collaborate more effectively, and ultimately make better decisions.
The data-access problem behind slow reporting
Stripo: There's often a disconnect between what email marketers actually need from martech vendors and what vendors think they need. Having sat on both sides, client-side and as a founder, where do you see that disconnect most clearly?
Kevin: I think vendors sometimes build features because they're technically impressive, while customers buy software because they want fewer problems.
One thing I've learned is the value of staying close to your clients. Because we focus on a very specific audience — CRM and email marketers — we have the advantage that our customers share many of the same challenges. That makes it much easier to build our roadmap with them rather than making assumptions from behind a desk.
The larger your audience becomes, the harder it becomes to understand everyone's needs in detail. I've often seen large software suites where the person you're speaking with isn't actually familiar with your day-to-day challenges.
I believe that software companies should spend less time building features that sound exciting in launch announcements and more time solving the problems their customers face every day.
Stripo: SEINō's core promise is helping marketers structure their data and analyze results faster. How exactly do you help with this, and how does it improve the lives of email marketers?
Kevin: One of the biggest frustrations for marketers is not only analyzing data but also getting access to it in the first place.
Campaign data sits scattered across your ESP, your CRM, and sometimes a data warehouse, and stitching it together usually means opening a ticket with IT or waiting on a data team's backlog. We built SEINō so that marketers could connect their own data and start asking questions directly, with no ticket required.
The difference between SEINō and a generic BI tool is that SEINō already understands the CRM language: campaigns, sends, clicks, conversions, revenue, audiences, benchmarks, and targets. It also comes with specialized AI skills and analytical tools built specifically for CRM and email marketing.
The real shift isn't that reporting gets faster. On the contrary, it's that marketers stop waiting for someone else to hand them an answer. They explore their own numbers, build a report in the time it used to take to request one, or just ask a question in plain language and get an answer back. That's the direction we're building toward: less time waiting on data, more time acting on it.
From channel to identity: Where email is really heading
Stripo: Looking ahead: between Gmail and Yahoo's sender requirements, the steady erosion of third-party data, and AI reshaping both creation and inboxing, where do you think the email channel itself is heading over the next few years?
Kevin: I think one of the biggest shifts isn't just happening within email itself, but around digital identity.
Email has become one of the strongest identifiers we have across digital systems. Even if people increasingly communicate through messaging apps, AI assistants, or entirely new interfaces, email often remains the underlying identity that connects customer data across platforms.
At the same time, stricter sender requirements and increasing privacy expectations will continue to raise the quality bar. The organizations that succeed won't necessarily be the ones sending more emails, but the ones sending more relevant, trusted, and personalized communication.
Technology will keep changing, but the principle behind email won't. People still want useful, timely, and relevant conversations with brands they trust. I don't think AI changes that; it simply raises expectations for how well we deliver on it.
Wrapping up
Thank you to Kevin for sharing his thoughts on KPI frameworks and the role AI can play in analytics. One thing is clear from our conversation: data isn’t there to prove that your assumptions were right. Rather, its most useful job is disproving what you were sure of.
A simple way to put this into practice is to look at the last report you sent to a stakeholder. Mark each metric as something you own or something you observe. Anything you can't move belongs in your context section, not your results.
Good analytics should help you understand what's really happening and use that to make better decisions. The tools will keep changing, but you'll get the most from them by letting your data tell you when you're wrong.
0 comments