Contact UsOur BlogArticlesManaging $70K+/month in ad spend for Ontario businessesManaging $70K+/mo in ad spendService AreasPortfolioClient Login
← All posts

Ecommerce Marketing Automation

The Second Order Is Where a Store Becomes a Business

An online store owner writing up orders beside packed boxes, a laptop and a rail of clothing, the kind of business ecommerce marketing automation is built for

Ecommerce marketing automation earns its keep on the second order, not the first. Most stores set it up the other way round: an abandoned-cart email, a welcome discount, a weekly newsletter to everybody, and a dashboard that counts opens. Then they wonder why repeat revenue looks the same as it did before.

The flows are not the hard part. Any store platform will send a cart reminder. The hard part is deciding which customers are worth a message, what each product actually earns after costs, and when a particular buyer is likely to need the next thing. Get those three decisions right and the automation is a short list of flows that run for years. Get them wrong and you have built a machine for sending discounts to people who would have bought anyway.

We'll cover:

Ecommerce Marketing Automation Starts With Margin, Not Flows

Before you write a single email, work out what each product leaves you per order. Not the price. Not the return on ad spend. What is left after the cost of the goods, shipping, payment fees and returns.

That number decides almost everything downstream. A store usually finds its range is doing two jobs. A handful of products can carry the cost of finding a new customer. A much larger group only makes money when it is sold to somebody already on the list. Automation belongs mostly to the second group: the accessories, the refills, the second size, the thing that goes with what they bought.

It also decides what you can afford to give away. A 15% welcome code on a product with a 30% margin after costs is half your profit, handed to a customer on the order you most needed to make money on. Plenty of stores run that code for years because nobody did the arithmetic once.

Returns deserve their own column. In some categories they take most of the margin, and a flow that pushes more orders at the customers most likely to send things back will look great in the email report and terrible in the bank account. Count kept orders, not placed ones.

An afternoon with a spreadsheet does this. It changes which products get promoted in which flow, which customers get a discount at all, and how hard you can afford to chase a lapsed buyer.

Three Segments Cover Most of the Value

The single most expensive habit in store email is sending the same message to everybody. A customer who bought once last month and one who bought four times last year are different people. The message that suits one is noise to the other, and every piece of noise teaches the reader to ignore the next one.

You do not need a dozen segments to fix that. Three do most of the work:

  • Recent buyers. They have just trusted you with money. What they need is reassurance that the order is coming, help getting value from the product, and, at the right moment, the thing that pairs with it.
  • Lapsed buyers. They bought, liked it well enough, and drifted. A reminder of what they bought and what has changed since does more than a sale banner.
  • Started but did not finish. They got as far as a cart or a checkout and stopped. Often the reason is shipping cost, an account requirement, or a question about returns that the page did not answer.

Each needs a different message, and all three can be set up once and left running. That is the shape of good ecommerce marketing automation: a few deliberate paths, not a calendar of blasts.

Five Ecommerce Marketing Automation Flows, in the Order They Pay

On the stores we work with, the flows we build are abandoned-cart, welcome, post-purchase, win-back and launch, automated in the CRM so repeat revenue builds without more ad spend. The order matters. Build them in this sequence and each one is paying before you start the next.

Abandoned cart and checkout

The buyer has already chosen the product, so the message is about removing the reason they stopped, not persuading them all over again. Answer the shipping question, the returns question and the sizing question in the email itself. Hold back the discount for the second or third message, and only on products whose margin can carry it.

Before you build this flow, look at the checkout. If shipping appears only at the last step, or the store insists on an account, the email is treating a symptom.

Welcome

A new subscriber has told you they are interested and not much else. Use the first few messages to show what the brand is for, which products people start with, and what makes them come back. Ask one question you will actually use, such as what they are shopping for, and write the answer to the record.

Post-purchase

This is the flow most stores underbuild and the one with the best economics. Confirm, then help: how to use it, how to care for it, what to expect. Then, at the point the first purchase would plausibly be running out or wearing in, offer the thing that pairs with it. A customer who has bought once is far likelier to buy again than a stranger is to buy at all, and reaching them costs almost nothing.

Win-back

Set the trigger from your own data, not a default. If most repeat buyers come back within 60 days, a customer at 120 days is drifting. Remind them what they bought, show what is new in that line, and stop after a few attempts. A list padded with people who will never buy again drags down every send.

Launch

New products go first to the people who bought the related one. A launch sent to the whole list is a newsletter; a launch sent to the right buyers is a sale.

Timed to the Product, Not the Calendar

A Tuesday newsletter is a decision about your schedule. A message that arrives three weeks after a purchase, when a 30-day supply is getting low, is a decision about the customer. The second one gets read.

Every product has a natural rhythm. Consumables run out. Clothing has a season. Furniture and equipment have a long gap and then a related purchase: the cover, the accessory, the matching piece. Write that rhythm down product by product, or category by category if the range is large, and use it to set the wait timers in the post-purchase and win-back flows.

Hands writing on a labelled shipping box beside bubble wrap, an order going out that an ecommerce marketing automation flow can follow up on

When ecommerce marketing automation should stay quiet

Frequency is the cost nobody puts on the spreadsheet. Every message that is not useful lowers the chance the next one is opened, and a list that has learned to ignore you is worth very little. Four genuinely useful messages a year beat a weekly newsletter nobody opens.

Build suppression rules alongside the flows. Someone who just ordered should not get the abandoned-cart email for the same item. Someone in the middle of a return should not get a cross-sell. A customer who has not opened anything in a year should stop receiving the regular sends before they mark you as spam. And in Canada, the record has to show consent before any commercial message goes out at all: CASL requires it, express or implied.

One Record Behind the Store, the Ads and the Inbox

Ecommerce marketing automation can only be as good as what it knows about the customer. If the order history lives in the store platform, the email history in a separate tool, the text messages somewhere else and the customer-service conversations in a shared inbox, every flow is working from a partial picture.

On the systems we build, every lead from every source routes into one CRM record, and the conversation runs as one stream across SMS, Messenger, Instagram DM, WhatsApp and live chat. For a store that means a single place that knows what someone bought, what they asked about, what they were sent and whether they replied.

That is what lets the suppression rules above work. It is also what lets a customer who messages about a delivery get an answer from somebody who can see the order. The same system can handle the parts of the lifecycle stores tend to bolt on separately: review requests after the order is marked complete, broadcast campaigns to smart-list segments, and reactivation sequences for customers who have gone quiet.

It often costs less, too. Consolidating CRM, email, SMS, scheduling, payments and reviews into one system usually means several separate subscriptions can be cancelled.

Let the Ads Know Who Already Bought

The flows and the ad accounts are usually run by different people, and they should not be strangers. Ecommerce marketing automation that stops at the inbox leaves half its value on the table. Paid social for a store works as a funnel: prospecting finds cold buyers, retargeting closes the browsers and cart-abandoners, and retention keeps existing customers buying. Each stage needs its own offer and creative, and the record decides who belongs in which.

Three connections pay for themselves quickly:

  • Keep buyers out of prospecting. Paying to show a first-order discount to somebody who bought last week costs you twice: the click, and the margin on an order they would have placed anyway.
  • Hand the abandoners to retargeting at the right moment. If the cart email has not worked after a day or two, an ad showing the same product is a second reminder through a different door.
  • Send the results back. Closed sales can be pushed back to the ad platforms as offline conversions, which lets them optimise toward revenue rather than clicks or sign-ups. Wire the Meta Pixel, the Google Ads tag and GA4 into the same system so the numbers come from one place.

There is a second-order effect here that changes how a store competes. A business that knows its customers buy three times can afford to pay more for the first order than a competitor who only counts one. Good retention automation is what makes that number true, and it shows up in every auction both stores enter.

How to Judge Ecommerce Marketing Automation

Open rates tell you a subject line worked. They do not tell you whether the automation made money. Judge it on the numbers that move the business:

MeasureWhat it tells you
Repeat rateWhether customers are coming back. The headline number for retention flows.
Average order valueWhether post-purchase and launch flows are adding the right second product.
Revenue per flow, store-verifiedSales reconciled against the store's own records, not the email tool's attribution.
Contribution margin per orderWhether the discounts in the flows are buying profit or giving it away.
Kept ordersOrders minus returns. A flow that drives returns is not driving revenue.

The reconciliation matters more than it looks. The email platform, the ad platforms and the store will each claim the same order if you let them. Pick the store as the source of truth for revenue, label anything else as platform-reported, and never add the platforms together and call it a total.

Finally, look at what the automation did not do. If a flow has been running for six months and repeat rate has not moved, it is not a flow that needs better copy. It is a flow aimed at the wrong customer, at the wrong time, or with an offer the margin cannot support.

The Bottom Line

The flows are the easy part. The work is knowing what each product earns, which customer is which, and when they will need the next thing. Decide those once and the automation runs for years; skip them and it sends discounts to people who would have bought anyway.

  • Work out contribution margin per product before building anything. It decides which products each flow promotes and who gets a discount.
  • Three segments, recent, lapsed and unfinished, cover most of the value. Stop sending everybody the same thing.
  • Build abandoned cart, welcome, post-purchase, win-back and launch, in that order, and time them to how each product is used.
  • Run the store, the inbox and the flows from one record, and tell the ad accounts who has already bought.
  • Judge it on repeat rate, order value and store-verified, kept revenue, not opens.

Want Your Store's Repeat Revenue Built Properly?

For product brands we run the creative, the funnel and the follow-up as one system: abandoned-cart, welcome, post-purchase, win-back and launch flows built in the same CRM as everything else, measured against store-verified revenue. Book a strategy call and you'll talk to Cruz or Keegan directly, not a sales rep, and leave with a written growth plan that is yours to keep whether we work together or not. See our e-commerce and retail playbook and how we build CRM automation, or contact us.

Related Content

Straight answers

FAQ

What store owners and product-brand founders ask before they build their flows.

What is ecommerce marketing automation?

It is the set of messages a store sends automatically in response to what a customer does: abandoning a cart, signing up, buying, going quiet, or buying something related to a new product. Done well, it is a short list of flows timed to how each product is used, run from one customer record, and judged on repeat purchases rather than opens.

Which automation should an online store build first?

Abandoned cart and checkout, because the buyer has already chosen the product. Then welcome, post-purchase, win-back and launch. Post-purchase is usually the most underbuilt and has the best economics, since a past customer costs almost nothing to reach.

Should every automated email include a discount?

No. Work out what each product leaves you after the cost of goods, shipping, payment fees and returns first. Many products cannot carry a discount at all, and a code sent to someone who would have bought anyway is margin given away for nothing.

How often should a store email its customers?

Less often than most do. Messages timed to a customer's purchase do more than a fixed weekly newsletter, and a few genuinely useful sends a year beat a weekly email nobody opens. Suppress people who just ordered, are mid-return, or have not opened anything in a long time.

Do I need SMS as well as email?

Not for every flow. SMS suits time-sensitive messages such as delivery updates and abandoned carts, as long as the customer has consented to it. What matters more is that email, SMS and messaging all run from the same record, so one customer never gets three versions of the same nudge.

Want This Working For Your Business?

One call. Your market, your margins, a written growth plan you keep either way.

Book a Strategy Call

You'll talk to Cruz or Keegan, not a sales rep. Leave your strategy call with a written growth plan: yours to keep, hire us or not.