• Resources
  • Case Studies
  • American Big Box Retailer Cuts API Abuse by 84%, Eliminating Billions of Malicious Requests Daily

American Big Box Retailer Cuts API Abuse by 84%, Eliminating Billions of Malicious Requests Daily

Category: API Attacks
04/03/22

84%

Reduction in malicious API requests per day
PS5

Article Contents

    The Challenge

    The client is a big box retailer with over 1,500 stores across the United States. Their eCommerce website generates revenues of over $15 billion annually, selling a wide range of products including high demand items like gaming consoles.

    The eCommerce site sells a vast product range and is built on a microservices architecture using several APIs; like most big box retailers, their product listing API provides information on each item they sell, including price, availability, and specifications.

    Adversaries were exploiting this API, feeding custom-written scripts into bots to access product information at scale.

    This high velocity of API calls was impacting customers browsing the site, both directly by clogging up the API and slowing down response times, and indirectly by facilitating other attacks, for example snatching the full inventory of high-demand products such as PlayStation 5 and Xbox Series X consoles within seconds.

    With the peak trading season imminent, the retailer needed to dramatically reduce scraper traffic to protect the availability of the service, optimize resources, and maintain excellent service to customers.

    The Dangers of Web Scraping

    Malicious web scraping can help competitors undercut prices, steal content, or collect information and resources to prepare other attacks. For example, bots aggressively scrape product pages and APIs many times each second, looking for an indication the product is available. Scalper bots then swoop in instantly and buy up all the stock before genuine customers can react, forcing consumers to buy the products from scalpers on secondary markets at an inflated cost.

    Adversaries use scraping to gather information facilitating account takeover attacks, from which they can steal personal information and payment details.

    Content scraping is also a means for criminals to clone websites, facilitating fraud and scamming users out of login or payment details.

    Even otherwise harmless scrapers like search engine crawlers and product availability trackers can damage websites if a high volume of requests is made in a short space of time, overloading the service, and impacting performance and availability.

    The Solution

    Netacea’s ingest engine captures and processes every request. As the client is one of the biggest retailers in the United States, volumes of traffic are extremely high, even before factoring in aggressive scraping activity. This equates to billions of requests captured each day, peaking at over 200,000 requests per second during the initial proof-of-concept phase.

    Netacea captured all API requests using a low friction, low latency integration via the client’s CDN. This meant no changes to their applications, or any physical or virtual hardware, were needed. These requests were then analyzed by Intent Analytics®, our real time threat detection platform, which categorizes traffic by its intent and makes recommendations to permit, block or investigate in real time.

    Using AI to Mitigate Evolving Threats

    Given the volume of requests hitting the API, the only solution fast and accurate enough is AI-powered bot mitigation.

    API attacks can be generic or written specifically to attack individual APIs. Generic attacks are often high volume, come from a single or small batch of origins (IP addresses, datacenters, etc.) and are low complexity, making them easy to detect. More sophisticated attacks seek to avoid detection by more closely emulating human behavior.

    Intent Analytics uses machine learning algorithms to continually analyze every single request and rapidly categorize all types of API attack, from broad to targeted, fast to slow, and simple to highly distributed.

    Netacea’s adaptive AI patches new vulnerabilities as quickly as adversaries introduce them. We continually assess how the client’s API is being called, highlighting malicious activity, and taking corrective action instantly. Our machine learning algorithms constantly evaluate every API request by asking three questions:

    How does the user profile compare to known bad actors on this platform?

    Based on previously seen bot activity, our machine learning algorithms compare all traffic to all other user interactions.

    How does this user compare to other users currently using the system?

    Using dynamic clustering to group similar users, Netacea’s AI platform spots when new clusters are created, highlights atypical behavior, and constantly re-evaluates what ‘normal’ looks like.

    Is overall activity unusual?

    By utilizing recurrent neural networks and analyzing what has happened before and recently, we can predict what should happen in the next few minutes and highlight unexpected activity.

    The Outcome

    Netacea’s API Security reduced daily requests to the API by 84% (over 10 billion requests) within weeks of implementation. Mitigating API attacks has protected the client against content and price scraping, and scalper bots, as well as reducing infrastructure requirements.

    API bot requests reduced for big box retailer

    Attack overview

    • High demand, low supply items like PS5 and Xbox consoles targeted
    • Billions of requests made daily during peak trading season
    • Over half of all requests detected as malicious and blocked

    The three most popular requests during the peak trading season were to listings for the PlayStation 5, PlayStation 5 Digital Edition, and Xbox Series X console. While this wasn’t surprising given the popularity and scarcity of these items, Netacea’s AI and Bot Experts team detected that half of these requests, on average, were made by automated bots with malicious intent.

    It is likely that this scraping activity was just the first part of a larger attack, most likely scalping for resale on secondary markets. By blocking these requests, Netacea cut off the attackers’ kill chain, preventing later threats from occurring.

    Block Bots Effortlessly with Netacea

    Book a demo and see how Netacea autonomously prevents sophisticated automated attacks.
    Book

    Related Case Studies

    US American Football cover art photo
    Case Study
    10/05/24

    “The Big Game” Streamed Seamlessly to Millions Thanks to Netacea

    Netacea protected a major streaming service from outages during a major livestreaming event, mitigating huge credential stuffing attacks.
    Pill
    Case Study
    04/04/24

    Netacea Keeps an Online Pharmacy Safe from Scraping Attacks

    Aggressive scalper bots were threatening the availability of a major online pharmacy at peak times. Find out how Netacea protects them against malicious automation.
    Shoe
    Case Study
    05/09/23

    Netacea Detects 11x More Bots Than Previous Bot Solution for Luxury Shoe Retailer

    Learn how Netacea helped a retailer of luxury shoe brands spot 11 times more bad bots than their previous solution, resulting in a 73% reduction in web traffic.

    Block Bots Effortlessly with Netacea

    Demo Netacea and see how our bot protection software autonomously prevents the most sophisticated and dynamic automated attacks across websites, apps and APIs.
    • Agentless, self managing spots up to 33x more threats
    • Automated, trusted defensive AI. Real-time detection and response
    • Invisible to attackers. Operates at the edge, deters persistent threats
    Book a Demo

    Address(Required)