# How to run modern email marketing for a Shopify store

> Email marketing is now a system, not a string of sends. The parts, the order to build them in, and what changed: privacy, sender rules, attribution and AI.

By The Wizel Team · Published 2026-11-19

Most stores run email as a string of sends: a sale here, a new arrival there, a welcome flow someone set up two years ago and nobody has opened since. Modern email marketing is a system instead. Each part does one job, and the parts feed each other. The list feeds the flows. Segments decide who gets which campaign. Measurement tells you what to change next.

**1-3%** of visitors buy on their first visit. Email is how a store reaches the rest

This guide maps the whole system for a Shopify store on Klaviyo, whether you run email for your own brand or for a book of clients. It covers what changed in the last few years, the parts of the system and the order to build them in, how much email should earn, and what it can't fix. Each part has its own chapter that goes deeper.

## Why email is the channel to get right first

Paid ads are rented. Every visit costs money, the price keeps rising and the platform can change the rules overnight. The list is owned. Once you have paid to win a subscriber, the next email to them costs close to nothing, so what the list earns drops almost straight to profit.

How many visitors become subscribers changes what each one costs. Take a store spending $50,000 a month on ads. If it turns 8-10% of its visitors into subscribers, each subscriber costs about $10-12.50. At 2%, the same spend buys them at $50 each.

That is why the order matters: fix the back end before you scale the ads. Better flows and campaigns raise what each customer is worth over time, and a store whose customers are worth more can afford to pay more to win them. Scale ads first and you burn cash faster as the cost of each new customer climbs.

## What changed, and why the old playbook stopped working

The basics of email haven't moved: send people things they want, at a rhythm they accept. What moved is how you prove it worked, who lets your mail through and what everyone else's emails sound like.

| What changed | What it means for you |
| --- | --- |
| Apple Mail Privacy Protection (2021) can load the tracking pixel without anyone reading the email | Open rate is a rough signal at best. Judge emails by clicks, orders and revenue per recipient. |
| Gmail and Yahoo set rules for bulk senders in 2024; Microsoft started sending non-compliant mail to Junk in May 2025 | SPF, DKIM and DMARC, one-click unsubscribe and a spam complaint rate well under 0.3% are the price of entry, not best practice. |
| Consent is per channel and has to be provable | Email consent is not SMS consent, and a purchase is not consent to marketing. Log who agreed, when, where and to what. |
| Attribution windows decide how much credit email gets | The same program can read about 40% of store revenue on default windows and 5-10% on a 1-day click window. Only a holdout shows what email caused. |
| AI made it cheap to write emails that sound like everyone else's | Specific products, real details and one brand's voice are what stand out now. |

## The parts of the system

Thirteen parts, in roughly the order a reader meets them. Each links to its chapter.

1. **The inbox.** Authenticate the domain, record consent, and stop mailing people who never open or click. [Getting into the inbox](/blog/email-deliverability-and-consent).
2. **The list.** Sign-up forms with a real reason to join, quizzes that ask what people want, and SMS where it earns its place. [Growing a list that buys](/blog/grow-an-email-list-that-buys).
3. **The flows.** The automations that fire on what a shopper does: welcome, browse and cart abandonment, post-purchase, back in stock, win-back, and the rules for when two of them want the same person. [The flows every store needs](/blog/email-flows-every-store-needs).
4. **The segments.** Who is engaged, who is a first-time buyer, who is slipping away, and who gets which send. [Segmenting by behavior](/blog/segment-email-by-behavior).
5. **The month.** How many campaigns, to whom, and a mix where not every email asks for money. [Planning a month of email](/blog/plan-a-month-of-email).
6. **Peak season.** The lead-up to Black Friday, the peak itself and the January that follows. [The peak season plan](/blog/peak-season-email-plan).
7. **The words.** Subject lines, one job per email, and a voice that sounds like one brand. [Writing emails people read](/blog/write-emails-people-read).
8. **The design.** Built for a phone first, with buttons a thumb can hit and pictures that look real. [Designing for the phone](/blog/design-emails-for-the-phone).
9. **The offers.** A written discount policy, campaigns that sell at full price, and order-value tactics that protect margin. [Offers that protect margin](/blog/email-offers-that-protect-margin).
10. **The numbers.** Which metrics still mean something, how email revenue squares with finance, and tests that give real answers. [Measuring email after Apple's privacy change](/blog/measure-email-after-apple-privacy).
11. **Retention.** The second order, replenishment, subscriptions, loyalty and reviews: where lifetime value is made. [Retention and lifetime value](/blog/email-retention-and-lifetime-value).
12. **Your kind of store.** A drop brand, a grocery basket and a sofa people research for weeks need different programs. [Email by store type](/blog/email-marketing-by-store-type).
13. **Running it for clients.** The weekly rhythm, approvals, reporting and keeping the retainer. [Running email for clients](/blog/run-email-marketing-for-clients).

## Going deeper on one part

Nine deep dives take one part further, far enough to build it.

- **The list:** [SMS beside email](/blog/sms-marketing-for-shopify) and [quizzes and preference centers](/blog/quizzes-and-preference-centers).
- **The flows:** [the welcome series, email by email](/blog/welcome-email-series-guide) and [the post-purchase flow](/blog/post-purchase-email-flow).
- **The month:** [launches and drops](/blog/product-launch-and-drop-emails) and [selling into other countries](/blog/international-email-marketing).
- **The design:** [the page an email click lands on](/blog/landing-pages-for-email-clicks).
- **Retention:** [reviews, loyalty and referrals](/blog/reviews-loyalty-and-referrals-email) and [failed payments and subscription churn](/blog/failed-payments-and-subscription-churn).

## The order to build it in

You don't need all thirteen parts on day one. You need them in an order where each one makes the next pay off.

1. **Authenticate the sending domain and set up consent capture.** Before the first campaign. Unauthenticated bulk mail can now bounce, not just land in spam.
2. **Put a sign-up form on the site** with a reason to join, and log consent per channel.
3. **Build the core flows:** welcome, abandoned checkout and cart, browse abandonment, post-purchase and win-back. They run for every shopper, every day, without you.
4. **Split the list by engagement** so campaigns go first to the people who open and click, and people who have gone quiet get less.
5. **Plan the first month of campaigns** before making any of them: the sends, the audiences and the job each email does.
6. **Measure on clicks, orders and revenue per recipient**, and run a holdout on your biggest flow to learn what it really adds.
7. **Then deepen:** a discount policy, retention programs, peak season, and tuning for the kind of store it is.

## How much revenue should email bring in?

The common rule of thumb is 30-50% of store revenue, with anything under 25% read as a program that needs work. Use that number with care. It assumes Klaviyo's default attribution windows and a list that has had time to mature. On a 1-day click window, the same program can show 5-10%.

Attributed revenue also isn't the same as revenue email caused. In one practitioner's holdout test on a replenishment flow, 15.5% of the people who got the emails bought, against 15% of the people held back. The flow was credited with hundreds of thousands a month. What it actually added was about $7-8K. Flows that reach people who were going to buy anyway look far better in the report than in a holdout.

> Attributed revenue is what email captured. A holdout tells you what email created. Report the first, prove the second.

None of this makes email small. Its job is also to keep the brand in mind, and some of that value shows up later as orders through ads or direct visits that no email window credits. The point is to know which number you are quoting. The [measurement chapter](/blog/measure-email-after-apple-privacy) shows how to set up a holdout, and the [guide to attributed revenue](/blog/email-attributed-revenue-explained) explains what Klaviyo counts.

## What email can't fix

Agency operators put most of retention, 80-90% by their estimate, down to the product and how it is positioned. Email keeps the brand in mind, cross-sells and tests offers. It can't make people want a product they didn't like the first time.

- **Big jumps in repeat rate usually come from product:** a launch, a new flavor or colorway, a collaboration.
- **Churn can be an acquisition problem.** Pointing ads at the kind of customer who stays can do more than any retention flow.
- **Diagnose upstream.** If the ads, the offer, the sign-up form and the flows are all sound and the store still isn't profitable, look at product, positioning or price.

## If you run email for clients

An agency runs the same system once per client, and the hard parts move. Planning has to happen early enough for a client to approve it. Every week has to show the client something they can see. Reports have to compare clients fairly, since a drop brand and a grocery store will never have the same numbers. The [chapter on running email for clients](/blog/run-email-marketing-for-clients) covers the weekly rhythm, approvals and keeping the retainer, and [comparing stores fairly](/blog/compare-email-performance-across-stores) covers the reporting.

Start with the inbox: nothing else works if the mail lands in spam. Read [chapter 1, getting into the inbox and staying there](/blog/email-deliverability-and-consent).

## Set up a modern email marketing program for a Shopify store

1. **Authenticate the sending domain**: Set up SPF, DKIM and DMARC for the domain you send from before the first campaign, and confirm one-click unsubscribe is on.
2. **Capture subscribers with consent**: Add a sign-up form with a clear reason to join, and record consent separately for email and SMS.
3. **Build the core flows**: Set up welcome, abandoned checkout and cart, browse abandonment, post-purchase and win-back flows in Klaviyo.
4. **Segment by engagement**: Create segments by how recently people opened, clicked or bought, and send campaigns to engaged people first.
5. **Plan the month**: Lay out the month's campaigns, audiences and the job of each email before making any of them.
6. **Measure what matters**: Track clicks, orders and revenue per recipient rather than opens, and run a holdout on the biggest flow.
7. **Deepen the program**: Add a discount policy, retention programs, a peak season plan and tuning for the kind of store.

## FAQ

### What is modern email marketing?

It is email run as a system: an authenticated sender with consented subscribers, automated flows that react to what shoppers do, segments that decide who gets each campaign, a planned monthly calendar, and measurement on clicks, orders and revenue rather than opens.

### How much revenue should email drive for a Shopify store?

A common rule of thumb is 30-50% of store revenue, with under 25% read as underperforming. That assumes default attribution windows and a mature list. On a 1-day click window the same program can show 5-10%, so check the window before judging the number.

### Which email flows should a store set up first?

Welcome, abandoned checkout and cart, browse abandonment, post-purchase and win-back. They run for every shopper automatically and cover the moments closest to a purchase.

### Are email open rates still useful?

Only as a rough, secondary signal. Apple Mail Privacy Protection can load the tracking pixel without anyone reading the email, so judge emails by clicks, orders and revenue per recipient.

---

Source: Wizel (https://www.wizel.ai/blog/how-to-run-modern-email-marketing)
Last updated: 2026-11-19