Most lists of ecommerce email flows give every store the same five or seven automations. Some stores need all of them. Many need two or three done well, and the rest can wait until there are enough visitors and customers to make them matter.
This guide starts with your buying cycle, then covers welcome and purchase recovery, post-purchase, replenishment and win-back, and finishes with how to tell whether a flow is adding anything. The flow maps and emails below are examples to adapt, not templates to copy.
Map the buying cycle before picking flows
An email flow is an automated sequence triggered by something a person does: signing up, leaving a checkout, buying, or going quiet. It is only useful if the trigger happens often in your store and the message helps at that moment.
So before choosing flows, answer four questions about your customers.
-
How long do people take to buy the first time?
A $30 impulse purchase and a $1,200 piece of furniture have very different gaps between first visit and order. Longer decisions give welcome and browsing follow-up more work to do.
-
Do they need to buy again, and when?
Consumables run out on a rough schedule. Durable goods may not be replaced for years. That decides whether replenishment is a flow at all.
-
What stops them at checkout?
Delivery cost, timing, sizing, trust in a brand they have not bought from. Recovery emails work when they answer the real reason.
-
How many people hit each trigger?
A flow that fires ten times a month is worth less effort than one that fires five hundred times. Check volume before you build.
| Buying pattern | Example products | Build first | Can usually wait |
|---|---|---|---|
| Consumable, regular repurchase | Coffee, candles, pet food, skincare | Welcome, checkout recovery, post-purchase with replenishment timing | Browsing follow-up; win-back until you know the cycle |
| Considered, rarely repeated | Furniture, eyewear, prams | Welcome with buying guidance, checkout recovery, post-purchase care | Replenishment (often not relevant); frequent campaigns |
| Gift-led or seasonal | Jewellery, hampers, homewares | Welcome, checkout recovery, a dated reminder before the next occasion | Replenishment; short win-back windows |
| New store, small list | Any | Welcome and checkout recovery | Everything else, until the triggers have volume |
Examples to adapt, not a prescription. Your own order data should decide the order.
Low repeat purchase is not automatically an email problem. If customers do not come back because the product disappointed, delivery was slow or the range has nothing for them next, a win-back flow sends reminders to people who have already decided. How to get more customers to buy again covers those causes.
One rule applies to every flow. In Australia, commercial email needs consent, must identify the sender and must include a working unsubscribe, under the Spam Act 2003. Check what your signup and checkout forms actually collect consent for before you trigger anything from them.
Welcome and purchase-recovery priorities
For most stores, the first two flows are the welcome sequence and checkout or cart recovery. They reach people at the moments closest to a first purchase, and they are the flows that fire most often while a store is still building its customer base.
The welcome flow
A welcome flow starts when someone joins your list. Its job is to deliver what the signup promised, then answer the questions standing between this person and a first order. If you offered a discount, send it straight away. If you did not, send something useful instead.
| Example timing | Job | Stop the flow if | |
|---|---|---|---|
| 1. Welcome | Straight after signup | Deliver what the signup promised, say who you are in two lines, point to one starting product | They buy |
| 2. The main question | About two days later | Answer the question that stops most first orders, such as which roast suits their brewing method | They buy |
| 3. Proof | About four days later | Show reviews, how you roast and dispatch, and your delivery and returns terms | They buy |
| 4. A simple choice | About a week after signup | Narrow the range to two or three starting options | They buy, or have not opened any email |
Example timings to test, not a rule. The right gaps depend on how long your customers take to decide.
Subject: Your first bag, and how to choose it
Preview: Tell us how you brew and we'll point you to a roast.
Hi [first name],
Thanks for joining. We roast in small batches each week and post within two working days of roasting, so your coffee arrives fresh.
Most people ask us the same thing first: which bag should I start with? It depends on how you make coffee. If you use a plunger or pour-over, start with our medium roast. If you use an espresso machine, start with the house blend.
Not sure? Reply with how you brew and we'll suggest one.
[Your name], founder
Checkout and cart recovery
A recovery flow starts when someone begins a checkout, or adds to cart, and leaves without buying. Checkout abandonment is usually the stronger signal, because the person got further and you are more likely to have their email address and consent. Cart abandonment can only be triggered when your platform can identify the visitor.
The first reminder rarely needs a discount. Many people leave because they were interrupted, wanted to check delivery cost or were comparing options. Answer those first. A discount sent automatically to everyone teaches customers to leave carts on purpose, and it costs margin on orders that would have come anyway. Abandoned cart emails covers sequence design, timing tests and suppression in detail.
Subject: Your order is still here
Preview: Delivery times and returns, in case that's what you were checking.
Hi [first name],
You left [product name] in your checkout. We've saved it for you.
If you were checking the details: orders over [amount] ship free, most metro orders arrive in [number] working days, and you can return anything unopened within [number] days.
Questions about grind or roast? Reply and a real person will answer.
Post-purchase, replenishment and win-back options
Once someone has bought, the job changes. You are no longer persuading them to try you. You are making sure the first order goes well enough that a second one is likely.
Post-purchase
Order confirmations and shipping updates are transactional messages. A post-purchase flow sits alongside them and helps the customer get value from what they bought: how to use it, what to expect and how to get help. Ask for a review after delivery, once they have used the product, not on the day they paid.
For considered purchases that are rarely repeated, the most useful follow-up is often a care guide, or an accessory offered at the point it becomes relevant, rather than a reorder reminder.
| Example trigger or timing | Job | |
|---|---|---|
| Brewing guide | When the order is dispatched | How to store the beans and brew this roast well |
| Check-in | A few days after expected delivery | Ask how the first cups went and how to get help if something is wrong |
| Review request | About ten days after delivery | Ask for a review once they have used it |
| Replenishment reminder | Shortly before a typical bag runs out | Make reordering easy, and mention subscriptions if you offer them |
| Win-back | Well past their usual reorder time | Ask what changed, show what is new, and stop if they do not respond |
Examples to adapt. Timings should come from your own delivery times and repurchase data.
Replenishment
Replenishment only works if the product runs out on a roughly predictable schedule. Work out the gap from your own orders: look at customers who bought twice and the typical time between their first and second order. In this example, if a 250g bag lasts about three weeks for one daily drinker, a reminder at around two and a half weeks for a single-bag order is a sensible starting point to test. Larger orders need a longer gap.
Subject: Running low?
Preview: Your last bag was [roast name]. Reorder in one click.
Hi [first name],
If you've been having a cup a day, your bag of [roast name] is probably getting low about now.
You can reorder the same roast below. If you'd like to try something different, our [second roast] is close in flavour, with a little more body.
Buying every few weeks? A subscription means you don't run out, and you can pause or cancel any time.
Win-back
A win-back flow is for customers who have gone quiet. Define quiet relative to your own cycle: for a coffee customer it might be two missed reorders, for an eyewear customer it could be years, and for some durable products win-back is not a sensible flow at all.
Keep it short, make it easy to say no, and stop sending to people who do not respond. Continuing to email people who never engage can affect how inbox providers treat the rest of your mail.
Measure usefulness and commercial contribution
Email platforms report attributed revenue: sales from people who opened or clicked a flow email within a set window. That number is useful for seeing which flows are active. It is not the revenue the flow created. Many of those customers would have bought anyway, because they were already heading to the checkout or were loyal customers due to reorder.
The closest you can get to the real contribution is a holdout: randomly keep a share of eligible people out of the flow and compare how often each group buys. Some email platforms support this directly; with others you split the audience yourself. The answers can be very different from the dashboard.
| Measure | No discount in the flow | Flow includes a 10% code |
|---|---|---|
| People who received the flow | 1,800 | 1,800 |
| Orders from them (6.0%) | 108 | 108 |
| Orders expected without the flow (4.5%, the holdout group's rate) | 81 | 81 |
| Orders the flow added | 27 | 27 |
| Attributed revenue reported | $8,100 | $7,290 |
| Contribution the flow added | $1,012.50 | $202.50 |
Hypothetical figures for illustration. $75 average order before any discount and 50% contribution before media. Holdout of 200 people, 9 orders. The discount column holds order numbers constant to isolate the cost of the code; in practice a discount would also change behaviour.
The report says $8,100. The flow added about $1,013 of contribution. If it hands a 10% code to every buyer, it added about $203, because most of the code's cost goes to people who would have bought anyway: 81 orders × $7.50. That does not make the flow bad. It makes it worth much less than the dashboard suggests, and it tells you the discount is the first thing to test.
Keep sample size in mind. The holdout here was 200 people with 9 orders. A few orders either way would change the answer a lot, so repeat the comparison over several months, or use a larger holdout, before treating it as settled.
Revenue is not the only measure of whether a flow is useful. Also watch:
- Unsubscribes and spam complaints, flow by flow and email by email
- Replies, and the support questions a flow answers or creates
- Review volume after the review request
- Whether the second-order rate improves for customers who received the post-purchase flow
- The cost of discounts on flow orders, not just how often codes are used
If you want help deciding which flows your store needs and building them, email flows and campaigns are part of our ecommerce email marketing work, prioritised by what a Validation Sprint finds rather than a standard package. If you are building them yourself, an ecommerce marketing plan you can actually run helps you fit email alongside everything else competing for your time.
You are on the list. The next one lands on Friday.
About the author
Mareike NiedermeierMareike Niedermeier is the founder and CEO of Sales Savvy Online and the creator of Survivable Distance, a theory of ecommerce viability. Since 2019 her team has worked with more than 200 ecommerce brands. The Australian Business Journal named her a Top Ten Marketer in Australia.
More about Mareike →