Shopify email marketing: what to send, when to send it, and how to test
The campaigns every Shopify store should run (welcome, abandoned checkout, winback, weekly), the timing data behind them, and the two-subject-line test that improves every send.
Shopify email marketing works when two things are true: you send the campaigns buyers actually respond to, and you can see what each one earned. Here is the campaign set we run for every store, the timing behind it, and the testing system that makes every send a little better than the last.
The four campaigns that carry a store
1. The weekly campaign (your baseline). New arrivals, restocks, promos, education. This is the heartbeat: once a week minimum for most stores, twice during holiday season. Every campaign gets custom-designed around your products, not built from a template.
2. Abandoned checkout recovery. The highest-converting email you will ever send, because the buyer already chose. Fire the first recovery email within an hour; add a second touch at 24 hours. Pair it with SMS inside the 48-hour window for the strongest recovery curve.
3. Winback. Customers who have not ordered in 60-90 days get a “we miss you” with a reason to return: a restock of their usual, a new drop, a modest offer. This campaign runs itself and quietly produces the highest margin revenue, because acquisition cost is zero.
4. Welcome. New subscriber gets a short series: who you are, best sellers, social proof. List-growth pop-ups feed it; the customer intelligence layer segments buyers from browsers so the follow-up makes sense.
Order of operations matters when you build this stack: recovery first (it pays this week), winback second (it pays this month), welcome third (it compounds), weekly last (it needs a calendar habit). Stores that start with the weekly send and skip flows are doing the hardest version of the job first.
Timing that actually matters
- Send days: Tuesday through Thursday generally perform steadiest for retail; your own data beats every benchmark after 90 days of history.
- Send time: late morning local time is a dependable default; test morning against evening before you trust anyone’s universal rule.
- Recovery speed: abandoned checkout email inside 60 minutes; SMS inside 48. Speed beats cleverness here.
- Cadence discipline: one promotional send a week with a real reason beats three without one. Unsubscribes are the tax on boring.
One timing rule people get wrong: recovery windows. The first abandoned-checkout email at 20 minutes beats one at 4 hours, but the second touch should NOT follow an hour later. Space them: one hour, then 24 hours. The buyer who was interrupted needs the reminder; the buyer who said no twice needs you to stop.
The test that runs on every send
Every email we produce carries two subject lines. The campaign sends to a small slice of the list; after a few hours the winner by opens (or clicks, your choice) goes to everyone else. Over a year that compounds into a subject-line style guide built from your buyers, not a blog’s best practices.
What we see win repeatedly across accounts, so you can start smart:
- Specific beats clever: “Restock: the 30ml tincture is back” beats “Guess who’s back”
- Numbers in the line earn their place: counts, prices, deadlines
- The buyer’s own words beat the brand’s words: pull phrasing from reviews and support tickets
- Curiosity works once; it never works twice on the same list
Test one variable at a time. A subject-line test that also changed the send time tells you nothing you can reuse.
Attribution, or: was it worth it?
Every link in every email is rewritten to a tracked short link. Clicks are counted and forwarded instantly; orders are matched back to the campaign on an hourly sweep with a 7-day last-touch window. The result is a number per campaign: “$8,412” beats “email marketing works.” The full attribution pipeline is documented if you want the engineering detail.
The 7-day window is not an arbitrary choice: retail buyers click Tuesday and buy Friday, often after seeing the retargeting ad your email started. Same-day attribution models call that email a failure and you cut the program that was working. Match window to buyer behavior, not to what the dashboard defaults to.
Templates vs custom design
Template tools win on speed and lose on brand. The middle path that works: a designed layout system (your fonts, colors, product photography rules) that production fills per campaign, so every email looks like your store and ships in hours. That is how Trinacle’s email engine produces 4-8 campaigns a month on every Pro-and-up plan with human review on every send.
The tell that your template is costing money: customers who open but never click. Usually the template hides the products below a header nobody reads, or the buttons don’t look tappable on a phone. Design is a conversion variable, not a taste question: watch the click rate, not the compliments.
The compliance layer you cannot skip
Consent tracked per customer, one-click unsubscribe honored immediately, bounce and complaint monitoring with auto-pause guards so a bad send cannot torch your deliverability. Boring, non-negotiable, and included.
If you do this yourself this month
- Stand up abandoned checkout recovery first: it pays fastest
- Add winback at 60-90 days
- Pick one weekly send slot and keep it for 90 days
- Test two subject lines on every campaign, winner to the rest
- Install revenue attribution before you scale anything
Or have the team run all of it from $315/month, and read the pricing breakdown to pick your volume.