5 Email Flows to Fix Before Fall Traffic Arrives

Somebody built your welcome email on a Tuesday two years ago and dropped a 15 percent code into it. That code ran today. It ran yesterday. It ran on every full-price day you held on purpose this year, and it will run again tomorrow morning. Nobody approved it this quarter. Nobody launched it. It just never stopped.

Most stores audit the wrong thing. Once a quarter you open the promotional calendar, count the sales you ran, and decide whether you went too deep. That review covers maybe thirty days of the year. Your five core flows cover all 365. So your email flows are your real discount strategy. Not your marketing calendar. Your promo calendar is a window you open on purpose and close on purpose. Your flows are the thermostat on the wall behind you, holding a setting somebody chose once. Nobody walks past a thermostat.

By the end of this you will know which of your five flows spends the most. You will know how to prove whether any of them earned a single extra order. And you will know what to change before the traffic lands. Start with the thermostat. It has been running the whole time you were watching the window.

Your Marketing Calendar Is Not Your Discount Strategy

Ask any merchant how much they discount and you get a campaign answer. Two sales in spring. One in July. Whatever ran for back-to-school. All true, all on a calendar, and none of it counts the codes inside your automations. A code hard-coded into a flow is not an email. It is a standing price, open to anyone who trips the trigger, on any day.

The Thermostat Nobody Reset

So do the sort. Open last month's discount report and put every coded order into one of two piles. Codes from a campaign you decided to run. Codes that came out of a flow. Two minutes, and almost nobody has done it.

Here is a worked example. Use your own numbers. A store does 800 orders a month at an $84 average order value. Of those, 176 carried a code. The split comes back 78 campaign and 98 flow.

Now price it. Those 98 orders at $84 come to $8,232. A 10 percent code takes $823 off, in a month with no planned promotion in it. At a 45 percent gross margin, that markdown equals the gross profit on roughly 22 more full-price orders. Call it a leak and you will go looking for a hole. There is none. It is a second pricing policy, and it has never been to a planning meeting.

A campaign is a window somebody opened on purpose. A flow is the thermostat on the wall behind you, still holding a setting nobody has walked past in two years.

Five Dials, and You Probably Set Three of Them Wrong

Five flows matter. Welcome, abandoned cart, browse abandonment, post-purchase, and win-back. You already have all five. They work, the open rates are fine, and that is exactly why nobody opens them. In most stores at least three carry a code with no expiry date. Fall does not make a stale setting safer. A bigger, more motivated audience trips those triggers far more often than a quiet July did. The dial did not move. The volume tripled.

Cart and Browse Flows Discount the Highest-Intent People in the Store

Your abandoned cart flow triggers on behavior that predicts a purchase. Sit with that. Adding to cart is one of the strongest buy signals a session produces, and the standard fix is to discount it.

The Intent Inversion

A shopper picked a size, added it, and reached the cart page. That is the profile of a dedicated buyer, and a real share of that group comes back and pays full price with no contact at all. The flow cannot tell them apart. In the same list sit walk-away customers who drifted off, and window shoppers who added an item to check the shipping cost. All three get the same 10 percent.

Then the flow books the returning buyers as recoveries. The store reads that as proof the discount works, and most years it goes deeper. Browse abandonment goes further in the wrong direction. It fires on people who did nothing but look.

What That 9 Percent Is Actually Measuring

A recovery flow reporting a 9 percent conversion rate is not converting 9 percent of abandoners. It is collecting credit for every abandoner who bought inside its window, including the ones who never opened the email. Open rate cannot separate those groups. Click rate cannot either. The dashboard was built to report the flow, not to doubt it.

One honest measurement exists, and it is cheap. Withhold the flow from a random 10 percent of abandoners for four weeks. Compare purchase rate between the two groups. The gap is your flow. Everything above the gap was always coming back. Merchants skip this because the flow already looks like it works, which is the trap: your best-performing recovery flow is usually the most expensive thing you own.

Withhold the flow from ten percent of abandoners for four weeks. Whatever that group buys anyway is what your recovery discount has been paying for.

Timing makes the targeting worse. The email lands 45 minutes to four hours after the tab closed. By then the decision is made or gone, and your code arrives to congratulate whichever one happened.

A cart flow discounts everybody because it only ever sees one data point. The cart was left. It never saw the twenty seconds before that, where the difference between a dedicated buyer and a walk-away customer shows up. Growth Suite reads the session while it is still running. It scores likelihood to convert from behavior, not from one event after the fact. An offer shows only for visitors whose signals say they are leaving without buying. Everybody else comes back at full price.

The Other Three Flows Were Written for a Season You Are No Longer In

Welcome, post-purchase, and win-back share one defect. They fire on a calendar. A calendar does not know what season it is, what your margin looks like now, or what you sell. It only counts days.

The Welcome Code Sets the Price for the Whole Relationship

Say you pick up a subscriber during a fall spike. Four minutes after signup you hand them 15 percent off. You just taught the biggest batch of subscribers you will collect all year that your email list is where the discount lives. Every full-price campaign after that reads as the exception, not the rule.

Check this on your own file tonight. Pull the twelve-month average order value of subscribers who came in through a discounted welcome flow. Put it next to the ones who came in without a code. In most stores the discounted group never catches up, and the gap is wider than the code was deep. That code did not cost you 15 percent once. It reset what that person thinks your products are worth.

Post-Purchase Pays Somebody Who Already Paid

"10 percent off your next order," three days after delivery. Read that from the customer's side. They have the product in their hands. They like it. They needed no persuasion at all, and you just told a person who paid full price on Monday what they should have waited for. It is the moment when money buys you the least. It is also the moment your flow spends it.

Win-Back Runs on a Template's Clock, Not Your Product's

Most win-back flows trigger at 60 or 90 days because that number came with the template. A candle burns down in about five weeks. A backpack lasts three years. Both get the same "we miss you" and the same code. One far too early. The other months after buying the replacement somewhere else.

A win-back flow set to 90 days is not a retention strategy. It is a thermostat setting from a season you no longer sell in, still spending margin every quarter. Same for the welcome code you picked when your cost of goods was different. The number was probably right the day somebody typed it. It has outlived the store it was typed for.

Fix the Permanence, Not the Email

None of this is an argument for deleting your automations. An incentive inside an email is fine. The damage comes from codes that never expire. That is the part you fix, and three lines cover all five flows.

The Three-Line Repair

1. Give every code a lifespan. Open each flow and find every code. If a code has no expiry, it is not a promotion. It is your price. Give it hours, not years.
2. Set one depth ceiling. A single maximum across every automation. No flow gets deeper than the deepest campaign you would sit down and approve on purpose.
3. Decide which flows carry money at all. Post-purchase and browse abandonment usually should not. Both pay before you have any evidence the customer needed paying.

That leaves welcome and win-back carrying a capped, expiring code. Cart recovery carries a reminder instead of a markdown, and the live session handles the visitors who were genuinely walking away. Notice what you did not do. You did not add a flow or rewrite a subject line.

Do It This Week, Not During the Traffic

Order matters here. Audit the codes this week. Start the holdouts next week, while traffic is still readable. Then leave the flows alone once volume shows up. A change made mid-spike hands you a number you cannot read in January. You get a good month and no explanation for it, which is how the same dial survives another year.

Two mechanisms make the repair hold on its own. Growth Suite's email capture trades the address for a unique, time-limited code instead of a permanent one. The list keeps growing. The standing discount does not survive the signup. Growth Links let a flow carry a branded URL that applies its own code and reloads the abandoned cart. No public code sits in a template, waiting to end up on a coupon site. Every code is deleted server-side when its timer ends.

The goal is not fewer emails. It is that no code in your store outlives the reason somebody created it.

Who Set That Dial, and When?

Your promotional calendar covers about thirty days a year. Your flows cover all 365. The calendar takes the credit and the thermostat sets the policy. Cart and browse flows put a markdown on the strongest buy signals in the store, and their reported conversion quietly includes everyone who was returning without them. Welcome, post-purchase, and win-back fire on calendars written in a season you already left. It is not a leak. It is a setting, and a setting only changes when a person reaches up and changes it.

Here is the two-minute version, before fall traffic arrives. Pull last month's coded orders and split them into campaign codes and flow codes. Then check the date on every code in the flow pile. Anything older than the last time you looked hard at your margins is a thermostat setting, not a decision. Most merchants find at least one code set by a version of themselves who is no longer in charge of pricing.

If your cart and browse flows send one code to everyone who left, Growth Suite helps you tell walk-away customers apart from dedicated buyers. It keeps the first response inside the live session. One real, time-limited offer, shown only to visitors whose behavior says they are leaving, with the code deleted on the server when the timer ends. So you stop paying for orders you already had, without discounting the shoppers who were always going to buy. It is free to install on the Shopify App Store, with a 14-day free trial.

Frequently Asked Questions