I’ve been in marketing long enough to watch entire specialties get automated out of existence. But what’s happening in email right now is different-and most people are completely missing it.
For years, the best email marketers had this almost sixth sense about subject lines. You’d watch them work and it felt like magic. They’d stare at two options, tap their finger on the desk, and say “this one” with total confidence. And they’d be right.
That intuition built careers. It justified six-figure salaries. It was the thing you couldn’t teach in a course or replicate with a playbook.
AI is killing it. But here’s what makes this interesting: the way it’s dying reveals something uncomfortable about what that “intuition” actually was all along.
The Performance Theater
Every marketing meeting lately has the same energy. Someone proudly announces their AI tool boosted open rates by 15%. There’s nodding. Maybe some light applause.
What nobody mentions: so did everyone else’s.
When every brand uses similar AI models trained on similar data, we’re all optimizing toward the same local maximum. Your inbox probably already feels like this-a wall of nearly identical subject lines, all algorithmically tuned to the same frequency.
The open rate arms race is real, and everybody’s bringing the same weapons.
But while everyone’s focused on that battle, something more valuable is happening in plain sight.
What the Machine Actually Sees
Here’s where it gets interesting. Modern AI email systems aren’t just A/B testing subject lines anymore. They’re synthesizing signals no human brain could hold simultaneously:
- Real-time competitive analysis of what’s hitting inboxes right now
- Micro-behavioral patterns (that 3-second pause before opening)
- Cross-channel sentiment from ads, website visits, support interactions
- Cultural context-current events, weather, economic mood
- Individual timing preferences down to the minute
A senior email marketer might track a dozen variables consciously. The AI is processing thousands.
And here’s the thing that makes me uncomfortable to admit: when that experienced marketer says “I just know this will work,” they’re not tapping into some mystical creative force. They’re running pattern recognition on data points they can’t consciously articulate.
The AI is doing the same thing. Just better.
The Intelligence Play Everyone’s Missing
Most companies treat AI email tools as efficiency plays. Faster testing. Better optimization. More automation.
That’s thinking too small.
Your email AI is running millions of micro-experiments on your market every week. It’s discovering what language resonates, what timing works, what emotional triggers land. That’s not email data-that’s market intelligence.
Think about it: if your email AI discovers that “community” language suddenly outperforms “exclusive” messaging, that’s not a subject line insight. That’s a market shift that should influence your product roadmap, your ad creative, your sales pitch, maybe even your M&A strategy.
But I’ve talked to dozens of marketing leaders, and almost none of them are connecting these dots. Email performance lives in the email team. Paid social lives somewhere else. Product is in another building entirely.
The companies that figure out how to synthesize these signals across channels-those are the ones that’ll pull away from the pack.
A Practical Example
Let’s say you’re running Facebook campaigns and email simultaneously. Your email AI notices that subject lines mentioning “time-saving” started outperforming “money-saving” language about six weeks ago. That shift happened fast.
Most companies would optimize their email and move on.
Smart companies would ask: what does this tell us about how our market’s priorities are shifting? Should we be testing time-efficiency angles in our Facebook creative? Should product be emphasizing speed over cost in the next release?
Your email AI just gave you a six-week head start on a market trend. But only if you’re set up to listen.
The Creativity Question
There’s this narrative in marketing that creativity is inherently human. Machines optimize, but they can’t create.
I used to believe that completely. Now I’m not so sure.
What we called creative intuition in email-that gut feeling about what will work-turns out to be largely pattern recognition we weren’t processing consciously. We were running computations in our heads and calling the output “instinct.”
The AI is doing the same computation. Just explicitly.
This doesn’t mean creativity is dead. It means we were wrong about where it lives. The truly creative work isn’t writing the subject line-it’s deciding what question you want the answer to. It’s knowing when to ignore what performs best because it erodes your brand. It’s connecting insights across domains in unexpected ways.
Those skills? Still deeply human. And more valuable than ever.
The Convergence Problem
Here’s what keeps me up at night: we’re about 18 months away from inbox homogenization at scale.
Every major brand implements AI optimization. They all feed roughly similar data into roughly similar models. The algorithms all arrive at roughly the same conclusions about what works.
Suddenly, every email in a given category sounds vaguely alike. The performance advantage evaporates because everyone’s optimizing toward the same thing.
The brands that win won’t be those with the best AI. They’ll be those who know when to deliberately ignore what AI recommends to preserve differentiation.
This is the new strategic skill: understanding when optimization creates commodity and when human judgment needs to override the algorithm.
At Sagum, we’ve built our reputation on innovation-whether that’s navigating TikTok’s emerging ad ecosystem or scaling profitable Facebook campaigns. The principle is the same with AI email: best practices will make you average. Strategic deviation creates advantage.
Why Most Companies Will Botch This
I can already see how this plays out. In the next two years:
90% of companies will implement AI email tools, see a 10-15% performance bump, and consider it a win. They’ll treat it as a point solution-a better way to do what they were already doing.
10% of companies will recognize this as a fundamental transformation that requires:
Actual Data Infrastructure
Your AI is only as good as what it can see. Most companies have data scattered across systems that don’t talk to each other. Email platform over here, CRM over there, ad performance somewhere else, website behavior in another tool entirely.
If your AI can’t synthesize these sources, you’re optimizing in the dark.
Cross-Channel Integration
Email insights should inform your paid media. Ad performance should influence email strategy. Website behavior should feed both. Most marketing orgs aren’t structured for this kind of integration.
The companies that figure it out will have a massive information advantage over those that don’t.
Creative-AI Collaboration
The best results come from humans using AI outputs as raw material, not finished product. The AI shows you what’s working. Humans figure out why it’s working and how to amplify it.
But most companies haven’t built workflows for this collaboration. Creative teams and data teams speak different languages. They sit in different meetings. They have different incentives.
Brand Voice Protection
Left unconstrained, AI will optimize toward generic high-performance language. It’ll strip your brand voice in favor of whatever gets clicks.
You need explicit guardrails: “We never use urgency tactics.” “We avoid these emotional triggers.” “We maintain this tone even if it costs us performance.”
Most teams don’t have these constraints defined, much less implemented in their AI systems.
The technology is the easy part. The organizational change is what kills most attempts.
The Contrarian Take
Despite everything I just said, here’s what I actually believe: AI makes human expertise more valuable, not less.
It’s just that the type of expertise that matters is shifting.
Basic email skills are commoditizing fast:
- Copywriting variations
- Simple segmentation
- Standard A/B testing
- Template optimization
These are now available to anyone with a $99/month tool. They’re table stakes, not differentiators.
But strategic thinking becomes exponentially more valuable:
- Knowing when to override AI to protect brand equity
- Translating email insights into business strategy
- Designing experiments that answer strategic questions
- Building systems that amplify AI insights across channels
- Recognizing signal versus noise in performance data
The email marketers who survive won’t be the best writers. They’ll be the best strategists who happen to work in email.
We’ve seen this pattern before. The Facebook ads managers who thrived weren’t those who could manually optimize campaigns fastest. They were the ones who understood audience psychology, competitive dynamics, and strategic insight extraction.
AI accelerates this trend. It doesn’t reverse it.
What Leaders Should Ask Right Now
If you’re evaluating your email approach, here are the questions that actually matter:
1. “What are we learning from email performance that should change our product roadmap?”
If the answer is “nothing,” you’re missing the point entirely. Your email system is having millions of interactions with your market. It’s discovering what resonates. These aren’t email insights-they’re market truths.
2. “How are we maintaining differentiation as optimization drives convergence?”
What elements of your brand voice are non-negotiable, even if AI suggests changing them? Where are you willing to sacrifice points of performance to stay distinctive?
These should be explicit decisions, not accidents.
3. “What skills are we developing that AI can’t replicate?”
Your team shouldn’t be getting better at writing subject lines. They should be getting better at strategic thinking, cross-functional synthesis, and turning data into decisions.
Technical skills are depreciating assets. Strategic skills compound.
Infrastructure vs. Tactics
Here’s the pattern I keep seeing: tactics are commoditizing while infrastructure is differentiating.
Anyone can buy an AI email tool. That’s tactics. It’s available to everyone at roughly the same price point.
But building a system where email insights flow into paid media strategy, where performance signals influence product decisions, where cross-channel data synthesizes into unified intelligence-that’s infrastructure. And infrastructure is hard.
It requires executive buy-in. Cross-functional collaboration. Data engineering. Process redesign. Cultural shifts.
Most companies will optimize their subject lines and call it innovation.
A few will rebuild their entire marketing intelligence apparatus with email AI as one sensor among many.
The performance gap between these approaches will be enormous.
What We’re Seeing
At Sagum, we’ve spent over $2 million on TikTok advertising in the past year. We’ve been scaling Facebook campaigns for over a decade. We built our reputation on innovation, not imitation.
And across our client base, we’re watching this play out in real-time:
The convergence is already starting. Brands in the same verticals are beginning to sound alike because they’re optimizing toward the same signals.
The winners are treating AI as intelligence, not automation. They’re asking “what is this telling us about market dynamics?” not “how do we boost open rates?”
The integration opportunities are massive and mostly untapped. Almost nobody is systematically connecting email insights to their broader strategy.
The next two years will separate point solutions from platforms. Tactical wins from strategic advantages.
The Real Question
AI in email marketing isn’t about sending better emails. That’s the surface game.
The real question is: are you building a learning system or just an optimization tool?
Most companies are building optimization tools because they’re easier to implement and show immediate ROI. You can have one up and running in a week. Performance improves. Everyone’s happy.
But the asymmetric returns go to those building learning systems-treating email AI as one sensor in a broader intelligence apparatus that informs creative strategy, media planning, product development, even business strategy.
The question isn’t whether AI will transform email marketing. It already has.
The question is whether you’re playing checkers or chess.
Checkers is optimizing opens and clicks. It’s tactical. It’s measurable. It’s also available to everyone.
Chess is building strategic intelligence infrastructure. It’s harder. It’s less obvious. It’s also where the sustainable advantages live.
In a market where everyone has access to the same AI tools, strategic architecture is the only real differentiator.
The brands that win won’t have the best AI. They’ll have the best questions for their AI to answer.
Most companies aren’t asking good questions yet. Which means the opportunity for those who do is extraordinary.
