The Real Reason B2B Emails All Sound the Same, And It Isn’t AI 

The obvious way to use AI in email marketing is also the wrong one, and it’s costing you engagement. 

In a world flooded with AI-generated content, being unmistakably human is your greatest advantage. The hard part is actually pulling that off without ignoring AI or leaning on it too hard. That’s what this series is about: exploring real ways to use AI to sharpen your voice instead of flattening it. We’ll walk through strategy, video and podcasts, blog posts and case studies, and social and email — one post at a time. 

I know why you’re reading this. You’ll tell yourself it was the headline — that you’re always looking for ways to sharpen a campaign. That’s not it. You’re here because your emails aren’t performing the way they used to, and you’ve run out of easy explanations for why. 

So you’re doing what every talented, diligent marketer does when the numbers won’t move. You’re going to the events. You’re downloading the guides. You’re following the right people on every channel and platform. And none of it’s working, which is the part that actually scares you — because you’re doing everything right, and “everything right” isn’t the problem. 

Here’s the uncomfortable part: there’s no universal fix. If there were, you’d have found it by now, because you clearly know how to look. What you’re dealing with instead is a handful of possible causes: 

• Maybe your emails haven’t changed format or layout in years, even decades. 
• Maybe you’re writing for a version of your audience that doesn’t exist anymore — the one from three job changes and two platform migrations ago. 
• Maybe your newsletter has six stakeholders and zero owners. Everyone can veto a change. Nobody can approve one. 
• Maybe the last time someone suggested an A/B test, a VP said, “I like what we have,” and that was the end of the conversation. 

Chances are it’s not one of these. It’s three of them, maybe more, quietly reinforcing each other, which is exactly why no single fix has worked. Somewhere in the last year you’ve probably run into someone insisting email marketing is dead. You know they’re wrong. Email still works — you can feel it working for other people’s inboxes every day. What you don’t have is a clear next move, and that gap between “I know this isn’t dead” and “I don’t know what to do about it” is a genuinely uncomfortable place to sit.

Here’s the part nobody says out loud: changing it is scary, even the parts that are fully within your control. It’s easier to keep sending the same email and hope the open rates crawl back up on their own. They won’t. Open rates don’t fix themselves, and waiting around for them to is still a choice—you’re just letting inertia do it for you instead of doing it yourself. 

How do I know? Because I’ve been there, making every excuse on that list sound reasonable to myself. I blamed slow news days for bad open rates. I told myself our subject matter wasn’t sexy enough to compete with whatever else was sitting in someone’s inbox that morning, or that our readers came to us for information, not entertainment, so why would tone even matter. I was wrong. Not about one of those things. About all of them. 

Here’s the good news: you’re trying to fix your email problem at exactly the right moment. Learning to use AI in email marketing starts with the work that kept you from writing your best emails: audience research, analytics deep dives, and the subject line testing you never had time to run properly. The audience research. The analytics deep dive. The subject line testing you never had time to run properly. All of that used to eat the hours you needed for the part that actually mattered. 

But I need to be straight with you: AI is not going to save a bad email. If your newsletter is already formulaic and safe, running it through AI won’t fix that — it’ll launder it. You’ll get something faster, cleaner, and just as forgettable, because AI is very good at producing more of whatever you feed it. Use it to write the whole thing for you, and you haven’t solved the problem in that list above. You’ve just found a more efficient way to keep having it. Use it the right way, in the right places, and it’s genuinely the thing you’ve been looking for — not because it replaces your judgment, but because it finally frees up enough of your time and attention to use that judgment where it counts. That’s what the rest of this is about: how to use AI in email marketing to make more room for the judgment, specificity, and voice your readers actually respond to.

How to Use AI in Email Marketing Without Losing Your Voice

Step 1: Know your audience, then write to one of them 

Most marketers skip this step without realizing it. Not because they think it’s unimportant, but because they’re convinced they’re already doing it. Ask them to describe their audience, and they’ll rattle off job titles and seniority levels. That’s not knowing your audience. That’s knowing their LinkedIn profile. Real audience knowledge means understanding what they’re dealing with right now. They don’t open emails out of loyalty — they’re looking for something specific, and if you don’t know what that is, you’ll write around it every time.  

This is where AI earns its keep. Feed it your notes, personas, past campaign data, plus context it can’t infer on its own — recent industry shifts, new tools your audience has adopted. It’ll build you a real picture of who you’re writing to and how best to speak to them. Expect something closer to a briefing than a bulleted persona sheet: the pressures your audience is under this quarter, the language they use versus the language that makes them tune out, the objections that come up before anyone even asks. That’s the raw material — not a draft, not a voice. 

Then close the research and do what AI can’t. Pick one real person from that audience and write the email to them. Have a conversation, not an announcement. The research gives you the material. The one person you’re picturing is what makes it sound human. 

Step 2: Add your fingerprints 

AI-generated email has a texture now, and readers feel it even when they can’t name it. You avoid that texture by writing something only you could write. Somewhere in the body, drop in one detail that could only have come from your actual week — something you noticed, a number from a call, a line a customer said almost verbatim. Then test it: could this sentence have been written last month or next month just as easily? If yes, sharpen it until the answer is no. 

If you’re not the one with the expertise — you’re writing on behalf of a founder, an SME, a technical lead — the fingerprint doesn’t have to be yours. It just has to be specific. Borrow it from them instead. A stat they mentioned in passing during a call. A phrase they used that never made it into any official deck. The offhand thing they said in a Slack message that was more honest than anything in the brand guidelines. You’re not inventing the expertise. You’re just refusing to let it get sanded down into something generic once it reaches the page. 

Step 3: Know your why 

Here’s my confession: I get a lot of marketing email, both professionally and personally, and most of it goes straight to the trash — plenty of it unopened. Not because the product is bad, but because the email wasn’t actually written for me, for my situation, right now. 

Before you touch the subject line, before you think about polish, ask yourself two questions: why am I sending this, and why does it matter to the person receiving it? Not “what am I promoting” — why does their day get even slightly better for having read it? 

If you can’t answer both in a sentence or two, you’re not ready to hit send. And once you can answer them, run one more check: does everything in the email actually point at that one answer? It’s entirely possible to have a crystal-clear why and still bury it under three competing asks — read this case study, book a demo, follow us on LinkedIn. Pick one. Cut the rest, even the ones you like. 

Step 4: Let AI argue with you  

Most people use AI to agree with them. They ask it to polish a draft, fix the grammar, make it sound more professional — and it does, because that’s what it’s optimized to do by default. Almost nobody thinks to point it in the opposite direction and ask it to disagree. That’s exactly why this step matters. You’re using the tool for the one thing almost no one else is using it for. 

Hand AI your draft email and ask it to read it the way a skeptical version of your recipient would — not as a proofreader, but as a critic. Where does the argument feel thin? What claim would a sharp reader push back on? Where are you assuming context they don’t have? 

It’s not going to feel good. Reading criticism of something you wrote never does, even when you know it’s coming from a tool with no opinion of you one way or the other. Sit with it anyway. The discomfort is the price of catching something now instead of a reader catching it later. 

This isn’t about doing whatever AI tells you. Decide what’s actually worth fixing, and make the fix yourself, in your own words. You’re also free to ignore notes that don’t hold up. The point is finding your blind spots, not outsourcing the edit. 

Step 5: Optimize your subject line and preview text with the help of AI 

Subject lines and preview text are the two elements standing between your email and the trash folder, and they’re also where ruts show up fastest. Write enough newsletters, and you start reaching for the same three constructions without noticing — the same rhythm, the same “Here’s why…” opener, the same safe hook. You can’t see your own rut from inside it. AI can, because it doesn’t have your habits. 

Feed it your finished email and ask for more options than you’ll ever need — ten for the subject line, ten for the preview text. Push for real distinction between them, not ten versions of the same sentence with different adjectives. Ask it to draw on email best practices, and to make an actual recommendation with reasoning, not just a list. If you have a direction in mind, say so — preview text that leads with a specific claim, subject lines that open a curiosity gap without tipping into clickbait. 

Then take the final say back for yourself. AI can widen the field. It can’t tell you which option sounds like you, and it doesn’t understand your subject matter or your reader the way you do — it’s pattern-matching against best practices, not speaking as someone who’s lived in this industry. Read every finalist out loud. Keep the ones that sound like something you’d actually say. Cut anything that sounds like it escaped from a blog post about subject line testing. 

Step 6: Use AI to find patterns in your email analytics 

Pulling data is a task. Understanding it is a skill. It’s entirely possible to have the numbers in front of you and still miss the pattern sitting inside them — a subject line style that quietly outperforms, a structure that keeps losing readers halfway through. Pattern recognition is one of the things AI genuinely does well. Using it here isn’t cutting corners. It’s handing a task to the tool built for it. 

Feed your email performance data into AI and ask it to find the pattern: which quotes, which structures, which subject line styles actually correlate with opens and replies — traced, not guessed at. Then push past the pattern and ask for a recommendation. What should you try next, specifically, based on what it found? 

This is the step where AI and human judgment do their most equal work. Industry best practices are a fine starting point, but they were written for someone else’s audience. The rules you build from your own data, tested and proven over time, are the ones that actually hold — and getting there takes both what AI can surface and what you decide to do with it. 

Step 7: Let AI find the second life in what worked 

Every so often, an email outperforms everything around it — a reply rate that catches you off guard, an open rate that sends you back to reread your own subject line. Don’t just note it and move on. That email is telling you what your audience wants, and email is rarely the only place they want it. 

Hand the winning email to AI and ask it to pull out the core insight — not the whole email, just the idea that made it land — and reshape that into a few formats: a LinkedIn post, a blog section, a podcast angle. You’re not repurposing the copy. You’re repurposing the thing that made people respond. 

Email Marketing Isn’t Dead. Your Approach to It Might Be. 

Email marketing isn’t dead. What’s dead is the old way of doing things — the “tried and true” practices that stopped being true years ago. Turning everything into a formula is easy. It’s a template you fill in instead of a message you actually write, and after enough repetitions, you stop noticing which one you’re doing. 

What’s actually changed is your audience. They can tell, almost instantly, when something was written for a segment instead of a person, and they’ve run out of patience for pretending otherwise. 

None of the steps above are complicated. Know who you’re writing to. Leave something of yourself in it. Know your why before you worry about how it looks. Let something push back on you before your reader does. That’s not a new formula. It’s the opposite of one — it’s what email looked like before formula got there first. 

You’ve run out of good excuses. Not enough time to know your audience — handled. Not enough time to check your own thinking — handled. Not enough time to find the pattern buried in six months of data — handled. That’s how to use AI in email marketing: give it the work that helps you understand your audience, then write the message only you can write. Write to one person. Say something true. See what happens. 

---

If you enjoyed this content and want marketing know-how delivered straight to your inbox, you’ll love our Marketing Minds newsletter. Sign up here>>