E-commerce

How to handle customer questions about counterfeits and copies

How to handle customer questions about counterfeits and copies

July 1, 2026

"I bought on your site, but the product looks fake to me." "I saw a copy of your brand on Amazon, what are you doing about it?" "How can I be sure that this Vinted is not a counterfeit?" Three tickets where a poorly handled suspicion of counterfeit destroys trust or opens the door to abusive refunds.

The e-commerce counterfeit support covers direct post-purchase doubt, marketplace reporting, pre-purchase verification, and legal escalation. Distinct from COA alone (#565): here, it's about fake copy products and brand reputation, not duplicate certificates.

This guide #567 covers policy FAKE-SUP, flow FK-1 to FK-8, and matrix FAKE-MAP. Primary CS counterfeit content. Future CS pair auth doubts bot (#568).

Summary

Why does counterfeiting generate sensitive tickets?

A counterfeit ticket mixes emotion, money, and reputation. The customer sometimes accuses falsely; sometimes they are right. A third party reports a copy on a marketplace without buying from you. Every improvised response can validate a fraudulent refund or ignore actual fraud.

Five typical frictions

  • Direct post-purchase doubt: official website customer thinks the product is fake

  • Marketplace reporting: copy spotted on Amazon, Vinted, Leboncoin

  • Pre-purchase verification: how to know if it is authentic

  • Third-party purchase: fake bought elsewhere, brand contacted

  • Dispute refund: refund demanded without proof of investigation

EUIPO estimates that 66% of young people have bought at least one counterfeit product without knowing it, reinforcing the need for clear verification channels (EUIPO, e-commerce counterfeiting 2026). Without FAKE-MAP, agents promise refunds or deny claims without a process.

FAKE #567 vs AUTHCERT #565, NUMED #563, second hand and bot #568

Five types of content, five distinct trust perspectives.

Quick matrix

#565 = COA document. #567 = fake copy product and marketplace reporting. Pipeline: bot #568 explain verify → #567 investigate refund legal.

Which fake_* typologies should be classified?

Eight counterfeit ticket scenarios.

Eight fake typologies

  1. fake_direct_doubt: official website purchase, product seems fake

  2. fake_marketplace_report: reporting of a marketplace copy listing

  3. fake_verify_prebuy: how to verify before purchasing

  4. fake_third_party: fake purchased outside official channel

  5. fake_refund_claim: refund demanded on suspicion of counterfeit

  6. fake_copy_similar: similar product, copy or inspiration

  7. fake_warranty_claim: after-sales warranty on non-authentic product

  8. fake_legal_escalate: threat of police complaint or lawyer

Tags: fake, counterfeit, authenticity_doubt, marketplace. FAKE-MAP-GROUNDED: refund investigate from map only.

How to structure the FAKE-MAP matrix?

The FAKE-MAP matrix documents the counterfeiting process by SKU or category for agents and future bot #568.

FAKE-MAP Columns

  • fake_sku_scope: SKU category or all premium

  • verify_url: official QR serial registry page

  • auth_markers: hologram packaging serial location

  • official_channels: site stores authorized retailers

  • investigate_steps: photos proof agent checklist

  • refund_if_confirmed: refund policy if direct confirmed fake

  • marketplace_takedown: legal ops reporting process

  • legal_contact: brand protection escalation

Sync legal ops, /authenticity page, AUTHCERT-MAP verify_url if overlap and helpdesk macros. Quarterly audit: reported listings processed.

FAKE-SUP policy in six rules

Six rules for counterfeiting support.

  1. FAKE-MAP-GROUNDED: verify refund takedown from map only

  2. SOURCE-CLARIFY: direct purchase vs third party vs report only

  3. VERIFY-OFFICIAL-ONLY: auth proof via verify_url registry only

  4. INVESTIGATE-BEFORE-REFUND: no refund without investigate_steps map

  5. MARKETPLACE-ESCALATE: report → legal ops takedown playbook

  6. LEGAL-ROUTING: threat of lawsuit → legal_contact map

Flow FAKE FK-1 to FK-8

Eight-step counterfeit agent flow.

  1. FK-1 Intake: fake_* intent + source purchase order_ref

  2. FK-2 Source classify: direct third_party marketplace report

  3. FK-3 FAKE-MAP: verify_url auth_markers investigate

  4. FK-4 Proof collect: photos serial listing URL packaging

  5. FK-5 Verify: registry verify_url status if serial provided

  6. FK-6 Classify: doubt refund report legal warranty

  7. FK-7 Respond: macro FAKE grounded map

  8. FK-8 Execute close: refund exchange takedown legal tag fake_resolved

SLA: fake_direct_doubt acknowledge within 4 hours, investigate_steps launched same interaction.

Essential FAKE macros

Four macro agents.

FAKE-VERIFY-01

« Authenticity verification: official page [verify_url]. Authentic markers: [auth_markers map]. Official channels: [official_channels]. Proof outside this channel is not valid. »

FAKE-SOURCE-01

« Purchase source: [direct site / third party / marketplace report]. [If direct: investigate_steps launched, photos required [list].] [If third party: limited CS verify guidance, no brand refund.] [If report: marketplace_takedown process legal ops.] »

FAKE-MARKETPLACE-01

« Listing report [marketplace URL]: forwarded to brand protection [legal_contact]. Processing time [sla_takedown map]. Please do not contact the seller directly. »

FAKE-REFUND-01

« Counterfeit investigation order [order_ref]: [status in progress / fake confirmed / authentic confirmed]. [If fake confirmed direct: refund_if_confirmed map.] [If authentic: auth_markers explained verify_url.] Delay [investigate_sla]. »

Edge cases: COA, luxury, refurbished, and price comparison

Five cases outside the standard macro verify.

Third-party purchase Vinted: FAKE-VERIFY-01 guidance only. Brand refund reserved for direct purchase confirmed fake investigate_steps.

Essential fake KPIs

Five FAKE steering metrics.

  • fake_ticket_rate: counterfeit tickets / SKU scope sales

  • fake_resolution_sla: direct doubt close investigate delay

  • fake_refund_confirmed_rate: confirmed fake refunds / direct investigations

  • fake_marketplace_takedown_sla: legal processed reporting delay

  • fake_false_positive_rate: confirmed authentic / direct investigations

Target: fake_resolution_sla under 72 hours on fake_direct_doubt with documented investigate_steps.

FAKE anti-patterns

Five common mistakes.

  1. Immediate refund without proof: INVESTIGATE-BEFORE-REFUND

  2. Denying without checking registry: VERIFY-OFFICIAL-ONLY

  3. Third-party purchase refund: SOURCE-CLARIFY

  4. Signal ignored: MARKETPLACE-ESCALATE

  5. Minimized legal threat: LEGAL-ROUTING

FAKE with Qstomy

Qstomy on Shopify: detect fake intent, FAKE-MAP RAG source classify, verify_url cite, proof collect workflow, handoff #568 bot tier 1 doubts.

Pipeline: #568 bot verify explain limits → #567 investigate refund takedown legal execute.

Explore AI support and request a demo.

Checklist, FAQ and going further

FAKE Checklist (8 steps)

  1. FAKE-MAP v1: verify markers channels investigate refund

  2. Policy FAKE-SUP: 6 INVESTIGATE-BEFORE-REFUND rules

  3. 8 fake_* typologies: helpdesk tags

  4. 4 FAKE-* macros: VERIFY SOURCE MARKETPLACE REFUND

  5. Page /authenticity: verify_url auth_markers public

  6. Legal ops playbook: marketplace takedown contact

  7. 45-min agent training: source classify scenario

  8. KPI Dashboard: fake_* section 9

FAQ

Difference AUTHCERT #565?
#565 = COA duplicate verify document. #567 = counterfeit product copy marketplace refund investigate.

Customer bought fake on Vinted?
FAKE-SOURCE-01 guidance verify only. No brand refund.

Immediate refund?
INVESTIGATE-BEFORE-REFUND. FAKE-REFUND-01 after investigate_steps.

Is Bot #568 enough?
Tier 1 verify explain limits. Investigate refund execute → agents #567.

Going further

This week: publish FAKE-MAP SKU premium, test FAKE-SOURCE-01 direct vs third-party scenario, measure fake_resolution_sla.

Enzo

July 1, 2026

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

Subscribe to the newsletter and get a personalized e-book!

No-code solution, no technical knowledge required. AI trained on your e-shop and non-intrusive.

*Unsubscribe at any time. We do not send spam.

Subscribe to the newsletter and get a personalized e-book!

No-code solution, no technical knowledge required. AI trained on your e-shop and non-intrusive.

*Unsubscribe at any time. We do not send spam.