How to Re-Engage Fashion Shoppers Who Did Not Buy
Interested shoppers leave signals of different strength. How to read try-on and browse intent, then route it back through site, email, SMS, and retargeting.
Most re-engagement programs in fashion are one segment and one offer. The segment is “visited, did not purchase, last 30 days.” The offer is 15% off. It goes out on a Thursday, it recovers a handful of orders, and it trains a slice of the list to wait for the next one.
The reason it underperforms is not the creative. It is that the segment contains four completely different people, and the program treats them identically.
To re-engage interested fashion shoppers, rank the signal before choosing the channel. A single product view belongs in the next on-site session; completed or repeated try-ons can support email or retargeting; an abandoned cart may justify opted-in SMS. Match the response to intent strength instead of sending one discount to every non-buyer.
The job splits into two halves that most teams collapse into one. First, read the signal accurately. Second, route it back to the shopper through the channel it deserves.
The signal is already in your store data. The work is routing it somewhere the shopper will actually see it. Editorial image in Classic Antla disposable-camera style.
Interest comes in tiers
Rank your non-buyers by what they did, not by when they did it.
Tier one: viewed a product. The weakest signal on the list. She may have been comparison shopping, bored, or bounced by mistake from an ad. Treat a single pageview as curiosity.
Tier two: engaged with the product. She opened the size chart, switched colorways, read the fabric detail, or generated a try-on. Something specific held her attention, and you know which garment it was.
Tier three: engaged repeatedly. She previewed the same coat in two colorways, or came back the next day and previewed it again. Treat repetition as stronger than a single interaction because the shopper chose to spend more time with the same garment.
Tier four: abandoned with intent. Item in cart, or a saved preview she never converted. She had made the decision and something interrupted it.
The gap between tier one and tier four is enormous, and a “did not purchase” segment erases it completely. Shopify’s ecommerce optimization guidance argues the same thing structurally: work the specific step where the loss happened, since a generic intervention at the wrong step just adds noise.
What the blanket discount costs
Sending one offer to all four tiers has three costs, and only one of them shows up in the campaign report.
You pay margin on the shoppers who were going to buy anyway, mostly tier four. You teach the responsive part of the list that patience is worth money. And you spend your one good re-engagement impression on the weakest signal, because tier one is almost always the largest group by volume.
Klaviyo’s benchmark reporting consistently shows triggered, behavior-driven messages outperforming batch sends on a per-recipient basis, which is a polite way of saying that relevance beats reach. Litmus research on email engagement lands in the same place from the creative side: relevance and rendering quality do more for engagement than send volume.
What a try-on session records that a pageview misses
A pageview tells you a URL was loaded. A completed try-on tells you considerably more, and it is the reason try-on data makes such a clean re-engagement trigger.
You learn which exact garment she evaluated, not merely which page loaded. When the implementation passes variant data, you can retain the selected colorway too. Preview count and repeat visits show how often she returned to the decision. Across a session, three structured blazers and no knitwear is a merchandising pattern you can act on.
The Antla try-on feed surfaces the last 72 hours of generations by product, which gives a merchandiser a current view of what shoppers are considering rather than only what they bought. Where merchants offer extra generations in exchange for an email, tier two and three signals can also arrive attached to an identity, which makes off-site routing possible.
In the network of 100-plus fashion brands on Antla virtual try-on, a completed preview is followed by conversion rates averaging 35% above the non-user baseline. The shoppers who preview and still do not buy are therefore the most interesting non-buyers in your database, not the least.
Four practical routes back
These are the four routes most Shopify fashion teams can use without inventing a new channel. Pick by signal strength, identity, and consent.
| Signal | Best route back | What the message should carry |
|---|---|---|
| Viewed a product once | On-site only, on her next visit | Nothing intrusive: surface the item where she left it |
| Engaged or previewed once | Email, if you have identity, otherwise retargeting | The exact garment, in the variant she looked at |
| Previewed repeatedly | Email with her considered item, size and stock context | Availability, fit detail, alternate colorway |
| Abandoned cart or saved preview | Email first, SMS only if she opted in | The item, and a frictionless way back to it |
On-site returning state is the underused one. She comes back, and the store still knows what she was looking at, with her previous preview available without re-uploading anything. No message sent, no consent needed, no cost per contact.
Email carries the most detail and is the right default for tier two through four once you have identity. Execution belongs in your ESP: Klaviyo flows for fashion try-on covers the build, and Klaviyo’s help center documents where custom properties and events land on a profile.
SMS is for tier four and explicit opt-ins only. It is the most interruptive channel you own and the fastest to lose. One message about a specific item she left behind is defensible. A weekly promotional habit is not.
Retargeting is the only route that works without identity, which makes it the fallback for anonymous tier two. Build audiences around engagement events rather than all site visitors so you stop paying to re-reach people who bounced in four seconds.
For choosing which items and segments to act on first, segmenting shoppers by try-on behavior is the practical companion to this table.
Where this job ends
Two adjacent problems get confused with this one regularly.
Dormant subscriber win-back is a different job. Those people are on your list and have gone quiet, which is a list-health question rather than an intent question, and email re-engagement with virtual try-on handles it properly.
Cart recovery mechanics are also separate. If your issue is that cart emails get opened and not clicked, the problem is usually the destination rather than the message, which improving abandoned cart email CTR takes apart.
What not to do
Do not contact a single browse. Tier one gets an on-site experience, nothing more. Emailing someone because a page loaded is how unsubscribe rates climb.
Do not stack channels on the same signal. Email plus SMS plus a retargeting frame for one abandoned coat reads as surveillance.
Do not lead with a discount on tiers two and three. These shoppers stalled on uncertainty, not price. Answer the uncertainty first, and keep the offer in reserve.
Do not reuse her generated try-on image anywhere public. Her image belongs to her. It can appear in a message addressed to her if your terms and privacy policy cover it. It never belongs in an ad, a landing page, or a social post.
Do not keep signals forever. Set an expiration window that reflects the product cycle. A preview for a sold-out seasonal item should not keep triggering messages after the decision is no longer useful.
Re-engagement questions worth settling
What counts as an interested shopper in ecommerce?
Anyone who left an intent signal without ordering. In fashion the useful tiers are a single product view, an engaged product interaction such as a try-on, repeated engagement with the same item, and an abandoned cart or saved preview. Each one predicts something different and deserves a different response.
How do you re-engage shoppers without offering a discount?
Answer the reason they stopped. For apparel that is usually uncertainty about fit or appearance, so a message built around the specific garment they were evaluating, with a way to see it on themselves again, does more than a code. Save discounts for the shoppers who genuinely stalled on price.
Which channel is best for re-engaging fashion shoppers?
It depends on signal strength and whether you have identity. Weak signals belong on-site, mid-strength signals belong in email or retargeting, and only explicit opt-ins with strong intent justify SMS. The channel should escalate with the evidence, never ahead of it.
Can you use a shopper’s virtual try-on image in re-engagement messages?
Treat a generated try-on image as customer data. Use it in a message to that shopper only when your consent language, terms, and privacy policy clearly cover that use. Do not place the image in advertising, social posts, or public pages, and have counsel review the policy for your market.
Continue with the channel guides
- Shopify fashion growth use cases for how re-engagement sits among the other merchant jobs
- Personalization in fashion email marketing for which personalization tactics are worth building first
About the author: Aaron started Antla because clothes never arrived looking the way the product photos promised, and he wanted shoppers to see themselves before they paid. He has strong opinions about brands that answer every signal with a discount code, and he shares them unprompted.
Before your next re-engagement send, split the segment into the four tiers and check how much of it is a single pageview. If the strong tiers are thin, the fix is upstream: add Antla to your store so the shoppers who are seriously considering a garment leave a signal worth acting on.