E-commerce
June 28, 2026
“I only want to return the serum from the set; the rest is fine.” Your agent hesitates: how much to refund? The bundle was 20% off. Loop displays an error. The customer follows up. Three days for a calculation that should take thirty seconds.
ShipBob points out that multi-SKU bundles generate more complex partial returns than single-product orders, between prorated refunds and restocking components (ShipBob, Shopify bundles 2026). Richpanel estimates the total cost of a return at between $10 and $65 depending on the category, even before any calculation error (Richpanel, returns 2026).
This guide #185 covers partial returns support: packs, bundles, and multi-product orders. No existing content covers this customer service angle. Distinct from generic return policy and global returns bot: here, we focus on item-by-item arbitration.
Summary
Why do partial returns block support?
A partial e-commerce return mixes logical pricing, promotions, stock, and diverging customer expectations.
Three CS Tensions
Unclear Refund: The customer wants the catalog price, but you must apply the bundle discount
Component Restocking: The warehouse receives 1 out of 3 items from the kit
Ambiguous Policy: Undocumented whole pack return vs. individual item return
Typical Volume
Skincare and supplement brands: 25 to 40% of returns involve a multi-line or bundle order (DTC ops estimate). Windborne notes that native Shopify does not auto-calculate partial bundles (Windborne, partial returns 2025).
Cost of the Back-and-Forth
Each "how much do I get back?" exchange adds 2 to 4 messages. See actionable responses (#178).
Convert over 2,000 customers on average per month with Qstomy.
The world’s 1st Shopify AI dedicated to customer conversion
Empowering 200+ e-commerce merchants
How does this topic differ from standard feedback?
Several guides touch upon returns. #185 focuses on multi-products and bundles.
Generic return policy
Return policy: deadlines, shipping costs. Here: rules by bundle type and prorated calculation.
Return bot
Bot return pre-qualification and bot returns: intent qualification. #185 defines partial business rules that the bot applies.
Pre-purchase bundles
PDP bundles guide: merchandising. Here: post-delivery customer service.
Promise #185
Bundle type matrix, refund calculation method, portal workflow, and macros by scenario.
What types of bundles and commands need to be mapped?
Audit your partial returns catalog before writing macros.
Four bundle architectures
Single bundle SKU: single order line, indivisible Shopify kit
Multi-SKU kit: multiple lines linked by discount code or bundle app
Multi-product cart: not a bundle but -15% cart promotion
BOGO / gift with purchase: free item or item at symbolic price
ReturnGO reminder
Single bundle SKU = whole return required in native portal unless configured otherwise in the app (ReturnGO, bundles 2025). Multi-SKU = line-by-line return possible.
Shopify tags
bundle_fixed, bundle_component, gwp_free, final_sale. Cross-reference with reason for return analysis.
How to choose between returning the entire bundle vs. a single item?
The partial return policy must be explicit by product type, not based on agent feeling.
Decision Matrix
Discovery box 3 miniatures: full return only if opened (hygiene)
3-step routine sold separately: partial OK, prorated bundle price
Value pack -25%: partial OK if 2+ items kept, otherwise recalculate discount
Single SKU bundle shipped sealed: full return only unless a component defect is proven
Loop configuration
Toggle Partial returns ON + Don't allow exchange on bundle SKU to force component returns (Loop, bundles 2026).
Customer communication
Return page: "You can return one or more items from your set. The refund takes into account the applied bundle discount."
How do you calculate a fair prorated refund?
The partial refund calculation avoids over-refunding and chargeback disputes.
Method 1: Proportional Prorating
Item refund = (item catalog price / bundle catalog sum) × bundle price paid. AfterShip calls this proportional prices (AfterShip, bundle returns).
Numerical Example
Set paid €72 (serum 40 + cream 35 + oil 25 = 100 catalog, -28%). Return of cream only: 72 × (35/100) = €25.20, not €35.
Method 2: Equal Price Split
Bundle €90 / 3 items = €30 per return. Simple but unfair if catalog values are very different.
Method 3: Capped Standalone Price
Refund the min(standalone price, prorating). Protects margin if a customer buys a bundle then returns the expensive item alone.
Shopify Manual
Admin > Refund > select lines > adjust amount. Note calculation in order note for audit.
Which workflow: self-service portal or manual?
The partial return portal must reflect your rules without an agent ticket.
Loop / AfterShip Workflow
Configure bundles in the admin app (parent SKU + components)
Enable partial + prorata method
Mark parent bundle SKU as non-returnable if multi-SKU
Test return of 1 component: credit amount shown = prorated
Restock sync: Loop does not restock partial, manual in warehouse
Gorgias Manual Workflow
Customer requests a partial return: agent verifies eligibility using section 4 matrix, calculates using section 5, creates return label for the item, refunds exact amount in Shopify, and tags partial_return.
Multi-line Corporate Orders
See bulk orders (#174) for partial returns of corporate gifts.
Target Processing Time
Self-service partial: immediate label, refund upon warehouse receipt or auto if Loop is configured. Manual partial: 24h SLA with amount confirmed in the first response to avoid follow-ups.
Which macros are supported by partial scenarios?
SAV partial return macros in MANO: mirror, amount, options, next step.
Partial authorized, portal
"Order #[X], Routine Éclat set. You can return one or more items here: [Loop link]. Pro-rata refund of bundle discount (e.g. cream alone ≈ €25.20 out of €72 paid). Label generated instantly."
Partial refused, sealed bundle
"This set is shipped under a single sealed SKU. Return possible only for the complete pack within 30 days. If a component is defective, send photos: [link]. We will assess for an exception." See fragile products (#182).
Multi-product cart, cart promo
"Partial return OK. Refund = line price minus cart promo share (-15% distributed). Estimated amount [Y] €. Confirm items to return: reply 1, 2 or 3."
Credit note vs refund
Offer +10% credit note if partial and stock OK. Credit note vs refund.
How to manage free gifts, BOGO, and cross-line promotions?
The bundle return edge cases generate the most poorly managed tickets.
Free gift with purchase
Rule: returning GWP = cancel benefit or refund face value if customer keeps main purchase. Tag gwp_clawback if value > €15.
BOGO 2nd at -50%
Return item paid at full price: keep promo item or refund BOGO prorated. Document in public policy.
BNPL partial refund
Partial shop refund ≠ scheduled payments updated immediately. Macro: “Refund of [X] € launched. Alma/Klarna will adjust within 3-5 days. Contact financier if there is discrepancy.” BNPL Support.
Partial exchange
Customer wants to exchange 1 item from the kit: process partial return + new order or draft order complement price difference.
Final sale in bundle
If a component is tagged final_sale: non-returnable even if other items in the kit are eligible. Bot and portal must gray out the line with policy explanation.
How to align warehouse and restock on a partial return?
The partial return restock breaks down if ops does not know which component is returning.
Warehouse brief
Return label: Component SKU, not parent bundle SKU
QA: Box seal open = repack or destock depending on policy
Inventory: +1 component, not +1 pre-assembled bundle kit
ShipBob recommendation
Specify to the 3PL: partial return = restock components, whole bundle = restock parent SKU. Quarantine if consumable product has been opened.
Mandatory Shopify Note
partial_return | items: SKU-A | refund: 25.20 | method: prorated prop | restock: manual component.
Which KPIs and partial return signals should you monitor?
Measure partial return performance beyond the overall return rate.
Leading KPIs
Partial tickets / total returns: benchmark 20-35% for bundle brands
Messages before partial resolution: target < 2 with configured portal
Difference between calculated refund vs customer expectation: disputes < 3%
Self-service partial / total partial: target > 60%
Lagging KPIs
Over-refund detected during audit (target 0). Cost of partial vs full return. Repeat purchase post-partial resolved within 24 hours.
Monthly review
Top 3 SKU partial return, 1 policy or calculation adjustment.
Alert signal
Same bundle with > 15% partial return rate: merchandising review (box set content, PDP description) before loosening policy. See return reason analysis by SKU bundle.
How does Qstomy prequalify partial returns?
Qstomy identifies bundles, calculates prorata, and redirects to the portal or agent based on rules.
Features
Multi-line order lookup: detects bundle tags
Live prorata calculation: estimated amount per item
Branch policy: partial OK vs. full required
Custom Loop link with pre-selected items
Amount dispute handoff: brief + detailed calculation for the agent
Quantified DTC Scenario
Skincare brand, 3,200 orders/month, 18% bundles, 22% partial returns (126/month), manual calculation, 4.2 messages/ticket. Qstomy partial deployment + Loop prorata + section 4 matrix. After 10 weeks: partial tickets -51%, self-service 68%, amount disputes -74%, messages/ticket 1.7, audit over-refunds 0.
Explore AI support, Shopify, request a demo.
Which operational playbooks should be launched this week?
Playbook 1: bundles inventory (2 h)
List all active bundles. Classify section 3. Shopify tag. Prioritize top 5 by return volume.
Playbook 2: matrix + calculation (1 h)
Write section 4-5 by type. Shared pro-rata spreadsheet for support. 3 numerical examples validated by finance.
Playbook 3: Loop config (3 h)
Import bundles, partial ON, test 5 real scenarios.
Playbook 4: macros (45 min)
Publish section 7 Gorgias. Roleplay partial refused + GWP.
Playbook 5: warehouse brief (30 min)
Document section 9. Share with 3PL. Test reception of 1 partial.
Useful links
A well-framed partial return costs less than a poorly refunded full return: the customer keeps what suits them, you keep the margin and the trust.
Enzo
June 28, 2026
