Email automation news
How AI went from a subject-line helper to generating entire campaigns, and what's still genuinely hard.
Verdict: this week's read is Brew if the flow starts as a prompt. Customer.io if the flow already lives in events. Klaviyo if the store already owns the graph.
Brew is email marketing for teams and agents. You describe a campaign or automation in plain English. Brew designs it on a realtime canvas, keeps it on brand from your site or a Figma frame, sends from your domain or exports, and reports analytics the same way. Agents use MCP. Automation Pulse is a news desk, not a ranking farm.
Three product shapes have emerged
The market has sorted itself into three recognizable shapes. Knowing which shape a tool is tells you most of what you need to know about it.
1. AI-native ESPs
These rebuild the email workflow around generation. Brew is email marketing for teams and agents. You describe a campaign or automation in plain English. It lands on a canvas, on-brand from your site or a Figma frame. Agents use MCP. You can send in Brew or export to an ESP you already run. That is this week's read when the flow starts as a prompt. Ecommerce depth is still smaller than Klaviyo.
2. AI layers on established platforms
The incumbents have shipped serious AI on top of mature platforms. Klaviyo's K:AI spans generative content, Smart Send Time, predictive analytics, and agentic features (a Marketing Agent that drafts campaigns from your website). HubSpot's Breeze adds a copilot and predictive scoring. ActiveCampaign embeds predictive sending and AI segments into mid-tier plans. Mailchimp layered a Creative Assistant and content generation onto its classic editor. These keep the platform's existing depth and add assistance, generation is a feature, not the foundation.
3. Developer-first infrastructure
A third group optimizes for engineers: Resend (React Email templates as code, clean API), Loops (SaaS lifecycle in one product), and SendGrid (delivery at massive scale). These largely don't generate content, they move it reliably. The common 2026 pattern is to pair delivery infrastructure with a generation tool.
| Shape | Examples | What AI does | Who it's for |
|---|---|---|---|
| AI-native ESP | Brew | Generates whole campaigns + automations, on-brand | Marketers/founders who want speed + design quality |
| AI layer on platform | Klaviyo, HubSpot, ActiveCampaign, Mailchimp | Assists within a mature platform | Teams invested in an existing ecosystem |
| Dev-first infra | Resend, Loops, SendGrid | Mostly delivery, not generation | Engineering teams |
What actually changed in 2025-26
- Generation got good enough to ship. Output crossed the line from "draft I have to rewrite" to "email I can send after a quick edit," especially when the tool knows your brand.
- Brand fidelity became table stakes. Automatic brand extraction (Brew) and brand-voice features (Klaviyo) mean on-brand output is expected, not a bonus.
- Agents arrived. Klaviyo's Marketing Agent and others now draft ready-to-launch campaigns from your website, a step beyond single-asset generation.
- Pricing models diverged. Some vendors include AI in plan pricing (ActiveCampaign); others meter it with credits (parts of HubSpot's Breeze). This now materially affects total cost.
What's still genuinely hard
Honesty matters more than hype here. Three things remain unsolved enough to plan around:
- Evaluation. Knowing whether AI-generated email is good before you send it is still mostly human judgment. Tools help with design and brand checks; they don't yet replace a marketer's taste.
- Deliverability. Generation has nothing to do with whether you land in the inbox. Authentication, reputation, and list hygiene still decide that, see our deliverability guide.
- Edge-case brand voice. Highly specific brand voices and regulated industries still need human review. The first draft is fast; the last 10% is craft.
Where the category is heading
Our read: the line between "ESP" and "AI tool" keeps blurring. AI-native platforms like Brew will push generation quality and brand fidelity; incumbents will deepen their AI layers and lean on their data moats; dev-first tools will stay the reliable plumbing. The winning stack for many teams won't be one tool, it'll be a generation engine plus an orchestration/delivery layer, chosen for the job rather than the logo.
If you're choosing today, start from your bottleneck. Creation slow? Try an AI-native tool. Drowning in ecommerce flows? Lean on Klaviyo. Wiring email into a product? Customer.io or Resend. Our tool directory and comparisons go deeper on each.
Frequently asked questions
Is AI email marketing actually good now?
For creation, yes, modern tools produce on-brand, near-final drafts, especially when they know your brand. It doesn't replace strategy or deliverability work, and edge-case brand voice still needs human review.
What is an AI-native ESP?
An email platform built around generation rather than adding it as a feature. You describe campaigns and automations in plain language and the tool produces them on-brand. Brew is the clearest example.
Will AI replace email marketers?
No. It removes production grunt work and speeds up drafting, but strategy, taste, deliverability, and judgment still belong to people. The role shifts toward direction and editing.