One of the most common questions we get is:

"How many emails should I be sending every week?"

And honestly, it's the wrong question.

The real question is: how many emails should each subscriber be getting?

Because the person who bought from you last week and the person who hasn't clicked in six months shouldn't be getting the exact same experience.

That's where tiered cadences come in.

Today we're breaking down how smarter brands are structuring email frequency in 2026, how AI fits into the equation, and why sending the same number of emails to everyone on your list is quietly costing you revenue.

Also inside:
→ Prime Day isn't just Amazon's holiday anymore
→ Your "winning" email won because it was the only one you sent.
→ The AI workflow Jimmy can't stop using
→ AG1 expands beyond the green powder with a standalone omega-3 line

Prime Day isn't just Amazon's holiday anymore

Prime Day isn't just an Amazon event anymore. It's one of the biggest buying moments of the summer, and shoppers are actively looking for products well beyond Amazon's marketplace.

Omnisend broke down 7 ways brands can turn that surge in buying intent into revenue, from teaser campaigns and VIP access to segmentation and post-Prime Day follow-up strategies.

Amazon spent millions building the hype, now's your chance to capture it.

→ Steal the playbook

*Sponsored

The tiered cadence playbook: how to structure email frequency in 2026

Most frequency advice on the internet still says "send 1 to 3 emails a week." That worked as a baseline a few years ago. It stopped being enough on its own.

In 2026, the brands actually winning the inbox have stopped picking a single send count. They build tiered cadences. Different frequencies for different segments, layered on top of automated flows that fire on behavior. The result is more revenue per subscriber and lower unsub rates, even at higher total volume.

Here's how to build that for your program.

Why "1 to 3 per week" stopped working

Three things changed the game:

1. Apple MPP gutted open data, so cadence decisions made off open rate alone are flying blind

2. Gmail and Yahoo tightened sender requirements, so a single bad month can tank reputation across the entire program

3. AI is quietly running inbox sorting now, including Gmail Promotions tab logic and AI summaries on iOS, which means real engagement signals matter more than ever

A flat "send 3 a week to everyone" approach doesn't survive any of those. Highly engaged buyers want more. Cold subs need less. Treat them the same and you bleed both directions.

The new mental model: engagement-tiered cadence

Instead of asking "how many emails should we send," ask "how many should each segment get."

Segment your list into three or four tiers based on real engagement signals:

• Active: opened, clicked, or purchased in the last 30 days

• Warm: engaged in the last 31 to 90 days

• Cool: engaged 91 to 180 days ago

• Cold: no engagement in 180+ days

Use clicks and purchases as your primary signal. Open data is too noisy post-MPP to lean on alone.

Assign cadence by tier

Here's a starting framework. Adjust based on your category and AOV.

This is the lever most brands aren't pulling. Active segments can absorb more email than a flat program ever sends them. Cold segments get a fraction of what they currently receive, which protects your sender reputation.

Where AI fits in

Three AI levers worth turning on this quarter:

1. AI send-time optimization. Omnisend Smart Sending, Klaviyo Smart Send, Mailchimp Send Time Optimization. The platform watches each subscriber's behavior and ships the email when they're most likely to engage. Lifts on flow emails are commonly 10 to 20 percent. Free in most ESPs.

2. Predictive engagement scoring. Klaviyo predictive analytics, Bloomreach, Insider, Ortto. These score each subscriber for churn risk and predicted next purchase date. Use the scores to size your tiers and time your win-back sequences.

3. AI-generated subject lines and preview text. Tools like Hoppy Copy, Phrasee, and built-in Omnisend AI. Generate variants for A/B tests, especially on your high-volume Active tier where cadence is highest.

What to skip: don't let AI auto-pick your overall send count. The model optimizes for opens, which is exactly the metric MPP corrupted. Frequency stays a human call.

A sample weekly plan

Here's what a typical non-promotional week could look like for a brand running all four tiers:

That's 5 sends to Active, 3 to Warm, 1 to Cool, 0 to Cold. Total volume looks high, and each subscriber still gets what fits their engagement.

Flows run independently on top of this: welcome, browse, cart, post-purchase, win-back. They drive 25 to 40 percent of email revenue for most ecom brands and should never get pulled back to "fix" cadence problems.

The cadence audit: 5 questions

Run this on your program once a quarter:

1. What percent of your sends are going to subscribers who haven't engaged in 90+ days? (Goal: under 20 percent)

2. What's your revenue per recipient by tier, and is it climbing on Active?

3. Are spam complaints under 0.1 percent across all sends?

4. Is your Gmail postmaster reputation rated High?

5. Are flows still pulling 25 to 40 percent of email revenue, or has something slipped?

If three or more of those answers are off, your frequency mix is the problem. Re-cut your tiers before you cut total volume.

Final thoughts

Email frequency in 2026 is a tiering problem. Solve it with engagement segments, AI send-time, and a quarterly audit. The "1 to 3 emails a week" rule was a useful baseline a few years ago. The brands compounding revenue today have moved past it.

Pick your tiers this week. Build the cadence per tier. Turn on Smart Send. Run the audit in 90 days.

Sender reputation, unsub rate, and revenue per recipient all move in the right direction at the same time.

Your "winning" email won because it was the only one you sent.

Every email marketer swears they "believe" in testing.

Almost none of them actually do it.

Not because they don't want to, but because building three versions of every campaign means three times the work.

Three headlines. Three heroes. Three offers. Three rounds of QA.

So one version wins by default. And you call it a strategy.

Here's the inconvenient truth: The subscriber who bought yesterday should not be getting the same email as the ghost who's ignored you for six months. But they are. Because you're shipping one email and crossing your fingers.

Paid media figured this out years ago.. Emails are still stuck in 2014.

Allan is bringing you into 2026.

Upload your headlines, images, copy, and offers once. Allan builds the best performing combination for every subscriber, automatically, and learns from every send.

More revenue per send, every send.

Stop pretending you test. → Start free at getallan.com

*Sponsored

Knowledge drop:

The magic of AI isn't in the prompt, it's in what comes next. Jimmy shares how he chains prompts together to turn customer reviews and interviews into messaging customers actually want to read.

DTC wins:

AG1 just launched AG Omega 3, a new fish oil supplement delivering 2,000mg+ of omega-3s per serving, with a separate CoQ10 version aimed at healthy aging. It's the latest sign that AG1 is evolving from a one-product company into a broader wellness ecosystem.

Annnnd that’s a wrap for this edition! 

Thanks for hanging with Chase and me, always a pleasure to have you here.

If you found this newsletter helpful (or even just a little fun), don’t keep it to yourself! Share ecomemailmarketer.com with your favorite DTC marketer. Let’s get them on board so they don’t miss next week’s drops.

Remember: Do shit you love.

🤘 Jimmy Kim & Chase Dimond

PS - Your next best customer might be reading this right now. Want in? Email Jimmy to sponsor this newsletter and more.

Love this newsletter but want to receive it less frequently? Let us know by clicking here!