Handling returns and disputes on Tutus: a practical guide

A buyer in Berlin opens a box containing a €240 Burberry trench coat shipped from a seller in Lyon. The listing described the coat as "excellent vintage condition, minor wear." Under the left lapel, however, runs a three-centimeter fray in the lining that was omitted from both the item description and the photo gallery.

On traditional peer-to-peer fashion platforms, this triggers a predictable nightmare: three days of hostile back-and-forth messaging, generic template replies from automated support desks, and an eventual coin-toss resolution that leaves one party feeling cheated. Tutus treats returns and disputes not as emotional customer service crises, but as deterministic state transitions within a transactional escrow protocol.

The Architecture of the 48-Hour Inspection Window

When a package is marked delivered by the courier API on the Tutus universe, a 48-hour countdown timer begins. The seller’s payout does not drop into their wallet; it remains locked inside WEVONE’s transactional escrow ledger. This delay is not an arbitrary banking hold—it is the functional perimeter of the transaction.

During these 48 hours, the buyer has two options: confirm the item matches its representation, which immediately releases the funds to the seller, or flag a dispute. If the buyer takes no action within 48 hours, the escrow engine assumes implicit acceptance and auto-executes the payout.

This design removes the common seller anxiety of buyers keeping an item for three weeks before claiming it was damaged. The boundary is clear, immutable, and tracked directly against carrier telemetry.

Categorizing Disagreements: Remorse vs. Discrepancy

To manage expectations, Tutus explicitly divides post-delivery issues into two categories: Fit & Preference versus Material Discrepancy.

1. Fit & Preference (Buyer Remorse)

Second-hand fashion carries inherent sizing risks. A tagged size 38 from 1994 does not cut like a size 38 from 2024. If an item matches the measurements and description provided in the listing but does not fit the buyer’s body or suit their personal style, this is classified as buyer remorse.

Tutus does not force sellers to accept returns for fit or preference. Instead, the buyer is provided a one-click mechanism to relist the item on Tutus without paying platform transaction fees on the second sale. The original seller receives their escrow funds without penalty.

2. Material Discrepancy (Listing Error or Damage)

A material discrepancy occurs when the physical reality of the garment deviates measurably from its digital record. This includes:

• Undisclosed physical damage (tears, broken zippers, moth holes).
• Unmentioned stains, discoloration, or altered hems.
• Incorrect brand attribution, fabric composition, or counterfeit status.
• Missing accessories included in the listing photos (e.g., detachable belts or hoods).

How Mia Mediates: Image Verification and Context Memory

When a buyer flags a material discrepancy, they do not enter a public comment argument with the seller. They open a structured dispute flow within the interface, upload high-resolution photos of the issue, and provide a short factual description.

At this stage, Mia steps in as the primary mediator. Rather than relying on human support agents to manually read hundreds of chat logs, Mia evaluates the dispute using context memory and visual matching:

1. Metadata Cross-Reference: Mia compares the buyer’s claim against the original listing text, seller disclosures, and raw metadata captured during item creation.
2. Visual Inspection: Mia passes the buyer's macro photographs through image embedding models trained to detect fabric distress, comparing them against the original high-resolution upload set. If the seller uploaded a high-res photo showing the lapel damage, the claim is instantly flagged as seller-disclosed, and the buyer's claim is denied.
3. Pattern History: Mia pulls the historic dispute rates of both accounts. A seller with 200 successful 5-star transactions faces a different baseline probability profile than an account opened three days ago that has flagged three consecutive shipments.

A Worked Example: The Lyon-to-Berlin Trench

Let’s return to the €240 trench coat shipped from Lyon to Berlin.

• Hour 0: Package delivered. €240 sits in WEVONE escrow. Timer starts.
• Hour 14: The buyer unboxes the coat, discovers the frayed lining, and opens a claim under "Material Discrepancy: Undisclosed Damage."
• Hour 14:02: Mia analyzes the claim. The seller’s original listing contained eight photos, none showing the under-lapel area. The text description stated "minor wear" but did not mention lining damage.
• Hour 14:03: Mia determines the claim is valid based on an undisclosed physical defect. The system halts the escrow timer and offers two deterministic resolutions:
  • Option A (Partial Settlement): The buyer accepts a €40 partial refund from the escrow hold to cover local tailoring repairs. The remaining €200 is released to the seller. Both parties must consent.
  • Option B (Full Return): The buyer receives a prepaid, trackable return label generated via Tutus logistics. The escrow hold remains active.
• Hour 18: The seller accepts Option B. The buyer drops the package at a Berlin logistics point. Upon the courier scanning the return package, the system automatically refunds €240 to the buyer, while return shipping costs are deducted from the seller's balance.

The WEVONE Mechanism: Contribution Scores and Escrow Health

Disputes do not exist in isolation on Tutus; they feed directly into the broader WEVONE platform architecture. Every transaction informs a user's platform standing.

When a dispute resolves against a seller due to negligence or omitted defects, it impacts their Contribution Score. A declining contribution score has practical platform consequences:

• Lower algorithmic priority in Tutus search results and feed curation.
• Extension of the standard escrow hold period on future sales (e.g., from 48 hours to 96 hours).
• Temporary suspension of instant payout privileges.

Conversely, sellers who consistently provide accurate descriptions, clear defect photography, and swift dispute resolution earn higher contribution scores, granting them lower platform fees and immediate escrow releases upon verified package delivery.

Real Limitations and Open Challenges

While structured state machines resolve 90% of routine fashion disputes, physical marketplace infrastructure comes with edge cases that technology cannot completely automate.

• Olfactory Disputes: Smells—specifically vintage mold, cigarette smoke, or heavy perfume—cannot be verified by image models. Currently, olfactory disputes require manual review by customer operations or mutual agreement on a flat cleaning allowance.
• High-Value Counterfeits: For items listed above €1,000, digital visual inspection is insufficient to guarantee authenticity. High-value transactions currently route through third-party physical verification nodes prior to final delivery, adding 48 to 72 hours to shipping times.
• Cross-Border Logistics Costs: Returning an item across European borders consumes economic value. On low-priced items (€20–€30), return shipping can exceed the value of the garment itself. We are currently testing automated micro-settlement models that offer buyers localized recycling credits rather than wasting transport energy on low-value returns.

By treating disputes as structured data problems rather than emotional battles, Tutus protects buyer trust without sacrificing seller security.