Your new junior marketer never sleeps: using AI for copy, design & analysis

Let's be honest: if your 2026 email strategy still involves staring at a blinking cursor for 45 minutes to write a subject line, you've already lost. AI has left the playground — it's a productivity lever that makes good marketers excellent and exposes lazy ones through generic mush.

We use AI daily — not to switch off thinking, but to execute faster. Here's our setup of external tools and the new gamechanger: the Klaviyo MCP server.

Part 1: the creatives — Claude, ChatGPT, Gemini

The most common mistake: "Write me a newsletter about our summer collection." The result is usually filler — clichés, no soul, no sales psychology. Use the tools like this instead:

Claude (the copywriter)

For copywriting, Claude is currently unmatched — it gets nuance and brand voice better than the rest. Use case: feed Claude your five best newsletters and say: "Analyze the style — tone, sentence length, humour. Then write an email on [product X] that addresses pain point [Y] in that style." Result: a draft that's 80% done and actually sounds like you.

ChatGPT (the strategist & coder)

Strong at structure and code. Use case (HTML): "Build an HTML scaffold for a Klaviyo section — image left, text right, a button that goes full-width on mobile, dark-mode compatible." Use case (ideation): "Give me 10 subject lines for an abandoned-cart flow. Audience: Gen Z. Tone: urgent but witty. No spam words."

Gemini (the researcher)

With direct access to Google data, it's great for context: "Which holidays or events next month are relevant for [dog owners]? Give me 3 campaign ideas."

Part 2: the revolution — the Klaviyo MCP server

Until now AI was "blind" to your data — it didn't know your last campaign's revenue unless you uploaded CSVs (tedious and questionable on privacy). Klaviyo solved it with the Model Context Protocol (MCP) server: connect Claude directly to your Klaviyo account, with read access to your live data — no more exports.

Instead of clicking through dashboards, you ask your account questions like an analyst:

1. Instant analysis: "Which of my campaigns had the highest ROI in the last 90 days — and why did it beat the average?"
2. Segmentation at speed: "Build a segment of customers who bought product A but not product B, with a customer lifetime value over €150."
3. Content audit: "Analyze all my flow subject lines from last year. Which words correlate with a low open rate?"

It democratizes data science — you don't need an Excel expert to find out why revenue dipped in November. You just ask your system.

Takeaway: AI doesn't replace strategy

Tools like the Klaviyo MCP server or Claude are powerful accelerators — but they're only as good as the questions you ask. If you don't know what customer lifetime value is, or why retention beats acquisition, the best AI just makes you faster at doing the wrong thing.

Want to professionalize your email marketing with AI without losing yourself in the tech? Talk to remantle Studio.