When I audit a store, the ad account is rarely the first place I find money. The email account is.
The pattern is almost always identical. There is one abandoned cart email, written when the store launched, still carrying a ten percent code. There is a monthly newsletter that goes out when someone remembers. And there is nothing else: no welcome sequence, no post-purchase sequence, no win-back. Meanwhile the ad account is spending steadily to bring in traffic that the store then has no way of talking to again without paying for the privilege.
That is the actual cost. Not the emails you are not sending, but the acquisition you are paying for twice because the first contact was never converted into a relationship.
Why this is an advertising problem, not a marketing one
Paid acquisition is priced per contact. Email is priced per month. When a visitor arrives from an ad and leaves without buying or giving you an address, the only way to reach them again is to buy them again — through retargeting, which costs real money, or through a fresh impression, which costs more.
So every flow below is really a way of converting expensive, rented attention into attention you own. The return on ad spend you see in the platform is not the number that matters. The number that matters is what a customer is worth over the following year, and flows are most of what moves it.
This is the same arithmetic as a healthy ROAS that leaves no profit behind, viewed from the other end.
The five flows, in the order I would build them
Build them one at a time. A rough flow running this week beats a perfect one scheduled for next quarter, because every day without it is a day of traffic you have already paid for walking away.
1. Abandoned checkout
Start here, always. These are people who entered payment details and stopped. They are the warmest audience you will ever have and the flow takes an afternoon.
Three emails. The first goes out within an hour, with no discount — most abandonment at this stage is distraction, a phone call, a nappy, a train stop, not price. Remind them what is in the basket, show the item, make the button obvious.
The second goes about twenty hours later and handles the real objection. Usually that objection is delivery: when it arrives, what it costs, what happens if it does not fit. Answer it plainly.
The third goes at around forty-eight hours and may carry a modest incentive, or free shipping, if your margin allows. Leading with a discount in email one trains people to abandon deliberately, and a surprising number learn that lesson quickly.
2. Abandoned cart and browse abandonment
Separate from checkout, and weaker, because intent is lower. Two emails is plenty. Browse abandonment — viewed a product, left without adding — is worth running only if you have enough traffic for it to fire meaningfully, and it should feel like a helpful reminder rather than surveillance.
Keep the tone light here. The further from checkout, the less entitled the email should sound.
3. Welcome sequence
This is the flow that converts a sign-up pop-up from a vanity metric into revenue. Anyone who joins your list from an ad campaign has raised their hand once; the welcome sequence is where you find out whether they meant it.
Four or five emails over a fortnight. Deliver whatever you promised immediately, in email one, with the code visible rather than buried. Then spend the next few telling them something worth knowing: how the product is made, what problem it solves, what the people who already bought say about it, what to buy first if they are unsure.
Sequences that sell on every email underperform sequences that sell on two of five. The job here is to earn the right to the inbox, and the sale follows from that.
4. Post-purchase
The most neglected flow and, for most stores, the most valuable.
A new customer is never more receptive than in the days after buying, and almost every store spends that window silent apart from a shipping notification. Use it. Confirm the order warmly, set expectations about delivery honestly, and then, after the item has arrived, explain how to get the best out of it.
A few days after that, ask for a review. Reviews from this flow compound into cheaper acquisition later, because social proof lowers the cost of persuading the next person.
Then, timed to the natural replenishment or complement point for your product, suggest the obvious next purchase. For consumables that might be four weeks; for a jacket, it might be an accessory a fortnight later. This is where repeat purchase rate actually comes from, and repeat purchase rate is what decides whether you can afford to outbid competitors on the way in.
5. Win-back
For customers who bought once and have gone quiet past your typical repurchase window. Two emails: one that assumes they simply forgot, one with a reason to return.
Win-back has the lowest conversion rate of the five and still beats cold acquisition on cost, which tells you most of what you need to know about the economics.
The things that quietly break flows
Sending every flow to everyone. If someone is in the middle of a welcome sequence and also triggers an abandoned cart, they should not get four emails in a day. Set exclusions so one flow suppresses the others.
Discounting at the first opportunity. Every incentive you hand out at the top trains the next purchase to wait for one. Use discounts where the objection is genuinely price, and nowhere else.
Leaving flows running unchanged for two years. Products change, shipping changes, the objections change. Read your own flows once a quarter as if you were a customer.
Measuring opens. Open rates have been unreliable since mail clients started pre-fetching images. Judge flows on revenue per recipient and on conversion, and ignore the rest.
Deliverability neglect. Authenticate your sending domain, warm it gradually if it is new, and remove people who have not engaged in a year. A beautifully written flow in a spam folder earns nothing.
What this is worth against the alternative
Across the stores I work on, well-built flows usually account for somewhere between a fifth and a third of total email revenue, and they do it without anyone pressing send. That revenue arrives at close to zero marginal cost, which means it lands directly on the bottom line rather than being eaten by auction prices that rise every year.
It also changes what you can afford to pay for a customer. If a buyer is worth meaningfully more over twelve months than they are on the first order, you can outbid competitors who only count the first order — and in a crowded auction, the business that can pay more usually wins.
If you are heading into a sale period, build the post-purchase flow before the sale rather than after it. The five-week Black Friday plan explains why that fortnight afterwards decides whether the whole exercise made money.
And if your checkout is where people are disappearing in the first place, flows are a patch rather than a cure — most abandoned baskets die on the same screen, and that is worth fixing first.
We build these as part of our Shopify development work, and the free growth audit covers the flows alongside the ad account so you can see which gap is costing more.