Short answer: Sidekick can build a full abandoned-cart recovery sequence and the Shopify Flow that actually sends it — copy, timing, discount code, and the live automation. Win-back is assembled slightly differently: you build the lapsed-customer segment in Shopify first, then run a sequence against it. Browse abandonment is the weakest of the three, because it depends on knowing who the anonymous visitor is.
That ordering is the takeaway: start where the automation is genuinely end to end, then layer on what needs manual setup.
Abandoned cart: the one Sidekick fully builds
Give Sidekick your brand name, typical cart value, what you sell, brand tone, whether you’re offering a recovery incentive, and how many emails you want. It produces, for each send:
- Timing after abandonment (e.g. 1 hour, 24 hours, 48 hours)
- A subject line built on curiosity or urgency
- Preview text
- Short body copy focused on overcoming objections
- CTA button text
Follow the escalation that actually converts:
- Email 1 — gentle reminder. Show cart contents. No discount yet.
- Email 2 — handle objections. Shipping, returns, trust signals.
- Email 3 — final nudge. Urgency, or the incentive if you’re using one.
Ask for dynamic placeholders for cart items and customer first name, and one trust-building element per email — reviews, a guarantee, or a support contact.
Then the part people don’t expect: tell it to build the automation too. Sidekick can create the Shopify Flow triggered by cart or checkout abandonment for each delay above, configure the trigger conditions and the send action through Shopify Email/Messaging, create the [DISCOUNT PERCENTAGE] recovery code with an expiry tied to the final email, confirm the Flow is on, and test it against a sample abandoned cart.
If you stop reading here, you already have the highest-ROI piece.
Win-back: build the audience before the emails
There’s no single “win-back” button. What makes win-back work is the segment underneath it, and that’s where Sidekick is genuinely strong — it can create real Shopify customer segments, not just describe them.
The segments worth having:
- Purchase-based: never purchased, one-time buyers, repeat customers (2-4 orders), VIPs (5+ orders or top spend tier), and lapsed customers — bought before, inactive since. That last one is your win-back list.
- Engagement-based: highly engaged, moderately engaged, and disengaged (no opens in 60-90 days). Define specific thresholds rather than leaving it vague.
- Behavioural: cart abandoners, browse abandoners, wishlist users, high-AOV versus discount-driven buyers.
- Demographic: product category interest, geography, acquisition source.
For each, ask Sidekick for the exact filter criteria, a recommended email frequency, the offer strategy, an example subject line approach, and — the underrated one — what not to send that group.
Then be explicit: create the actual Shopify customer segments now, using the filter criteria described, not just a written strategy. Without that instruction you’ll get a plan document instead of segments in your admin. Once the lapsed segment exists, reuse the abandoned-cart escalation against it: reminder, objection handling, incentive last.
Browse abandonment: real, but heavier
Abandoned browse targets people who viewed a product and never added to cart. The trigger design Sidekick will help you specify:
- Visitor viewed a product page but didn’t add to cart
- Must be an identified visitor — logged in, or email known from a previous visit
- Wait at least an hour before firing; they may still be browsing
- Don’t trigger if they purchased in that session
Plus smart conditions that keep it from becoming spam: skip anyone who purchased anything, skip unsubscribes, cap at one browse-abandonment flow per customer per week, exclude products under [MINIMUM PRICE] where the emails aren’t worth it, and if they viewed several products, feature the most expensive one.
The honest caveat: this is advisory work. Sidekick will design the sequence, write the templates, recommend the implementation route (Shopify Flow with Shopify Email, or Klaviyo), and specify the tracking you need — open rate, click rate, conversion rate, revenue recovered. But the identity requirement means coverage is inherently partial, and setup is more hands-on than abandoned cart.
Cross-sell: recover the customer, then grow the order
Once someone has bought, the follow-up email is where AOV moves. The important detail is that Sidekick doesn’t have to guess which products pair well.
Ask it to run a ShopifyQL query against your live order data to find what’s genuinely co-purchased with [RECENTLY PURCHASED PRODUCT] — the frequently bought-together items and their order counts — and populate the recommendations from those results.
Two email angles worth having on hand:
- “Complete the set.” Reference their purchase by name, explain why the additions complement it, show 2-4 products with prices and one-line descriptions, add a bundle discount if you’re offering one.
- “Customers also bought.” Social-proof led, showing what similar customers paired with the same purchase, with ratings or review counts next to each.
Keep it to 3-4 products maximum, and ask Sidekick to suggest send timing by product type — consumables and durables have very different reorder rhythms.
What to watch for
- Discount too early. Offering
[DISCOUNT PERCENTAGE]%in email one trains customers to abandon carts deliberately. - Overlapping flows. A customer can be a cart abandoner, a browse abandoner, and a lapsed customer at once. Set suppression rules or they’ll get three sequences in a week.
- Untested Flows. Always ask for the sample-cart test. A Flow that’s built but not enabled looks identical to one that’s working.
- Segments that never refresh. Re-check lapsed-customer criteria periodically, or your win-back list becomes people who already came back.
The bottom line
Abandoned cart is the piece where Sidekick genuinely goes from blank page to live automation in one session. Win-back is a segment problem first and a copy problem second, and Sidekick handles both — as long as you tell it to create the segments rather than describe them. Browse abandonment is worth designing, but budget real setup time. Build in that order and each layer pays for the next.