Unnoticed Reselling Hits D2C Right Before Golden Week. 1 in 23 E-Commerce Orders Fraudulent, Impacting an Estimated 134 Billion Yen Annually in the Domestic Market
Spider Labs reports an average fraudulent order rate of 4.38% in EC for 2025 based on an analysis of 7.34 million transactions. The company warns D2C businesses of the high risks of automated 'unnoticed reselling' ahead of Golden Week and urges pre-shipping security measures.
📋 Article Processing Timeline
- 📰 Published: April 24, 2026 at 18:00
- 🔍 Collected: April 24, 2026 at 09:31
- 🤖 AI Analyzed: April 24, 2026 at 09:39 (7 min after Collected)
Fraudulent purchases such as "sold out in seconds after release" and "resold at several times the list price" are not just problems for video game consoles and limited-edition goods. A survey by Spider Labs, Inc. (Minato-ku, Tokyo, CEO: Satoko Otsuki, hereinafter Spider Labs), provider of the anti-resale tool "Spider AF Resale Countermeasure" (hereinafter "Spider AF"), revealed that the average fraudulent order rate in e-commerce in 2025 is 4.38%.
Applying this to the approximately 3 trillion yen forecasted size of the domestic digital D2C market in 2025, it is estimated that transactions worth roughly 134 billion yen annually are under the influence of fraudulent orders.
Why Golden Week is a "Major Target Period" for D2C
For D2C businesses, Golden Week (GW) is a crucial period for acquiring new customers, heavily concentrated with first-time discounts, initial perks for subscriptions, coupon campaigns, and affiliate ads. Conversely, these campaigns are easily targeted by fraudulent purchasing groups, leading to tactics where first-time perks are repeatedly abused using multiple accounts or devices to resell products.
The problem is that these orders are difficult to distinguish from legitimate new customers. What is spreading in D2C is "unnoticed reselling" that goes undetected by both businesses and legitimate customers.
Since legitimate orders also surge during the GW sales period, manual visual checks and individual reviews tend to fall behind, making the risk of oversight particularly high.
Reference Metric for the GW Sales Period: About 1 in 23 EC Orders is Fraudulent - Analysis of 7.34 Million Transactions
As a result of analyzing approximately 7.34 million transactions across 98 accounts using Spider AF's fraudulent order detection function in live operation, the fraudulent order rate was 4.38% (about 881 cases per day). It is determined that roughly 1 in 23 cases flowing into EC sites is for purposes such as resale, mischief, or affiliate fraud.
*Survey overview is at the end.
Approximately 134 Billion Yen in the Domestic D2C Market is Affected by Fraudulent Orders
The size of the domestic "Digital D2C" market is predicted to reach 3 trillion yen in 2025. (*)
Applying the 4.38% fraudulent order rate detected by Spider AF to this market size, the transaction scale affected by fraudulent orders in the domestic D2C market is estimated at approximately 134 billion yen annually.
This is a scale that erodes the revenue structure of D2C businesses through the drain of first-time discount funds, apparent deterioration of subscription continuation rates (F2 conversion rates), fraudulent drain of affiliate rewards, and damage to brand value.
* Ureru Net Advertising Co., Ltd. "'Digital D2C' Market Trend Survey": https://www.ureru.co.jp/news/archives/122
3 Fraudulent Tactic Patterns to Watch Out For (From Spider AF 2025 Detection Results)
The main threats indicated by Spider AF's 2025 detection data are the following three patterns. The data center IP category accounts for about 33% of all fraud, and automated machine types like bots and crawlers make up the majority (about 52%).
1. Massive Bot Purchases originating from Cloud/Data Centers (approx. 50% of the total)
A tactic of making automated purchases via scripts from AWS, Azure, GCP, etc.
In 2025, concentrated orders exceeding 70,000 cases annually from a single IP were confirmed.
2. "Regional Spoofing" via Residential Proxies/VPNs (approx. 20-30%)
Orders routed through VPNs or residential proxies to disguise overseas access as domestic.
About 83% of fraudulent orders originate from domestic IPs, showing advanced disguise as legitimate access.
3. Targeting "First-Time Discounts" with Multiple Accounts/Devices (approx. 20%)
A tactic repeatedly using first-time discounts across multiple accounts, including "address jigging" (making minute changes to delivery info) and fraudulently obtaining affiliate performance rewards through self-made orders.
All of these are characterized by being difficult to distinguish from legitimate customers on an individual order basis and hard to catch with traditional IP restrictions or manual checks.
Spider AF's Analysis of 2025 Fraudulent Order Tactic Patterns
The Fortress Protecting the Brand is "Pre-Shipping Security"
Fraudulent reselling is a serious issue that can lead to damage to brand value, loss of fans, and social media backlash. Dealing with it after purchase is often too late, making the importance of finding and eliminating fraud at the pre-shipping stage ever greater.
As this analysis shows, about half of the fraud is automated attacks by bots and data center IPs. The remaining half is disguised as legitimate access using residential proxies and multiple devices, compounding tactics that are hard to catch with conventional IP restrictions.
EC companies that have actually implemented Spider AF have been able to counter "list-type reselling" that they couldn't detect on their own. As a result of continuously withholding shipments based on fraud detection data, the number of fraudulent orders decreased to one-third of the peak level.
Fraudulent Order Detection Service "Spider AF"
Applying this to the approximately 3 trillion yen forecasted size of the domestic digital D2C market in 2025, it is estimated that transactions worth roughly 134 billion yen annually are under the influence of fraudulent orders.
Why Golden Week is a "Major Target Period" for D2C
For D2C businesses, Golden Week (GW) is a crucial period for acquiring new customers, heavily concentrated with first-time discounts, initial perks for subscriptions, coupon campaigns, and affiliate ads. Conversely, these campaigns are easily targeted by fraudulent purchasing groups, leading to tactics where first-time perks are repeatedly abused using multiple accounts or devices to resell products.
The problem is that these orders are difficult to distinguish from legitimate new customers. What is spreading in D2C is "unnoticed reselling" that goes undetected by both businesses and legitimate customers.
Since legitimate orders also surge during the GW sales period, manual visual checks and individual reviews tend to fall behind, making the risk of oversight particularly high.
Reference Metric for the GW Sales Period: About 1 in 23 EC Orders is Fraudulent - Analysis of 7.34 Million Transactions
As a result of analyzing approximately 7.34 million transactions across 98 accounts using Spider AF's fraudulent order detection function in live operation, the fraudulent order rate was 4.38% (about 881 cases per day). It is determined that roughly 1 in 23 cases flowing into EC sites is for purposes such as resale, mischief, or affiliate fraud.
*Survey overview is at the end.
Approximately 134 Billion Yen in the Domestic D2C Market is Affected by Fraudulent Orders
The size of the domestic "Digital D2C" market is predicted to reach 3 trillion yen in 2025. (*)
Applying the 4.38% fraudulent order rate detected by Spider AF to this market size, the transaction scale affected by fraudulent orders in the domestic D2C market is estimated at approximately 134 billion yen annually.
This is a scale that erodes the revenue structure of D2C businesses through the drain of first-time discount funds, apparent deterioration of subscription continuation rates (F2 conversion rates), fraudulent drain of affiliate rewards, and damage to brand value.
* Ureru Net Advertising Co., Ltd. "'Digital D2C' Market Trend Survey": https://www.ureru.co.jp/news/archives/122
3 Fraudulent Tactic Patterns to Watch Out For (From Spider AF 2025 Detection Results)
The main threats indicated by Spider AF's 2025 detection data are the following three patterns. The data center IP category accounts for about 33% of all fraud, and automated machine types like bots and crawlers make up the majority (about 52%).
1. Massive Bot Purchases originating from Cloud/Data Centers (approx. 50% of the total)
A tactic of making automated purchases via scripts from AWS, Azure, GCP, etc.
In 2025, concentrated orders exceeding 70,000 cases annually from a single IP were confirmed.
2. "Regional Spoofing" via Residential Proxies/VPNs (approx. 20-30%)
Orders routed through VPNs or residential proxies to disguise overseas access as domestic.
About 83% of fraudulent orders originate from domestic IPs, showing advanced disguise as legitimate access.
3. Targeting "First-Time Discounts" with Multiple Accounts/Devices (approx. 20%)
A tactic repeatedly using first-time discounts across multiple accounts, including "address jigging" (making minute changes to delivery info) and fraudulently obtaining affiliate performance rewards through self-made orders.
All of these are characterized by being difficult to distinguish from legitimate customers on an individual order basis and hard to catch with traditional IP restrictions or manual checks.
Spider AF's Analysis of 2025 Fraudulent Order Tactic Patterns
The Fortress Protecting the Brand is "Pre-Shipping Security"
Fraudulent reselling is a serious issue that can lead to damage to brand value, loss of fans, and social media backlash. Dealing with it after purchase is often too late, making the importance of finding and eliminating fraud at the pre-shipping stage ever greater.
As this analysis shows, about half of the fraud is automated attacks by bots and data center IPs. The remaining half is disguised as legitimate access using residential proxies and multiple devices, compounding tactics that are hard to catch with conventional IP restrictions.
EC companies that have actually implemented Spider AF have been able to counter "list-type reselling" that they couldn't detect on their own. As a result of continuously withholding shipments based on fraud detection data, the number of fraudulent orders decreased to one-third of the peak level.
Fraudulent Order Detection Service "Spider AF"