How to Set Up Product Availability Monitoring on Marketplaces Using Proxies: A Guide for Sellers

Automatic stock availability monitoring is a critically important task for sellers and retailers. Tracking competitor inventory on Wildberries, Ozon, Amazon allows you to quickly adjust prices and purchases. But marketplaces strictly block mass requests from a single IP — without proxies, your parser will be blocked after 10-50 requests.

In this guide, we'll cover how to set up stable stock availability monitoring through proxies: which proxy type to choose, how to configure IP rotation, what request parameters to use, and how to avoid blocks on popular marketplaces.

Why Marketplaces Block Stock Availability Monitoring

Marketplaces protect their infrastructure from parsing with several levels of protection. When you launch automatic stock availability monitoring, your script makes hundreds or thousands of requests to the marketplace's API or web pages. For the platform, this looks like an attack or unfair data collection.

Main methods of parser detection:

• Request frequency from a single IP — if 100+ requests per minute come from one address, it's a clear sign of a bot. A regular user physically cannot open 2-3 product cards per second.
• Absence of cookies and session history — parsers often make requests without first visiting the homepage, which reveals automation.
• Identical User-Agent — if all requests come with the same browser header, it's suspicious.
• Behavior patterns — sequential enumeration of products by ID, absence of random pauses, perfect regularity of requests.

Blocking occurs at the IP address level. First, the marketplace may show a captcha, then temporarily limit access (rate limiting), and with systematic violations — completely block the IP for several hours or days.

Which Proxy Type to Choose for Stock Availability Parsing

The choice of proxy type directly affects monitoring stability and the likelihood of blocks. Different marketplaces react differently to various types of IP addresses. Let's consider three main options and their application for stock availability parsing.

Residential proxies — the optimal choice for most stock availability monitoring tasks. They use IP addresses of real home users, so marketplaces perceive requests as legitimate. For Wildberries and Ozon, this is practically a mandatory requirement — these platforms aggressively block datacenters.

Datacenter proxies are suitable for less protected platforms or when maximum parsing speed is needed. For example, for monitoring regional marketplaces or small online stores. The main advantage is high speed and connection stability. But on large platforms, they are blocked more often.

Mobile proxies — the most reliable but also the most expensive option. Use them for critically important products or when residential proxies start getting blocked. Mobile IPs are almost impossible to block, as thousands of real carrier users can be behind one address.

IP Rotation Settings: How Often to Change Addresses

IP rotation is a key parameter for preventing blocks. Proper configuration of address change frequency allows you to distribute the load so that a minimum number of requests come from each IP, imitating the behavior of regular users.

Two main approaches to rotation:

1. Time-based rotation
The IP address changes after a set time interval, regardless of the number of requests. This is a simple and predictable method suitable for regular monitoring with constant check frequency.

• Every 5-10 minutes — for frequent monitoring of critically important products (top positions, highly competitive products)
• Every 15-30 minutes — standard mode for most stock availability monitoring tasks
• Every hour — for background monitoring of large assortments when high update frequency is not needed

2. Request-based rotation
IP changes after a certain number of completed requests. This method gives more precise control over the load on each address.

• Every 10-20 requests — for Wildberries and Ozon (strict protection)
• Every 50-100 requests — for Amazon, eBay (moderate protection)
• Every 200-500 requests — for less protected platforms

Combined approach (recommended):
Set time-based rotation as the main method, but add a request limit as additional protection. For example: IP changes every 10 minutes OR after 50 requests — whichever comes first. This protects against random activity spikes.

Monitoring Specifics on Different Marketplaces

Each marketplace has its own parsing protection features and requires an individual approach to stock availability monitoring setup. Let's examine the specifics of working with popular Russian and international platforms.

Wildberries

Wildberries has one of the most aggressive anti-parsing protection systems among Russian marketplaces. The platform uses multi-level request verification, including analysis of headers, cookies, request frequency, and behavioral patterns.

Key features:

• Residential proxies are mandatory — datacenters are blocked almost instantly
• Full set of browser headers must be transmitted, including Referer and Accept-Language
• Wildberries API requires special tokens for some requests
• Frequent IP rotation (every 5-10 minutes) is critically important
• Add random delays between requests: 2-5 seconds

For stock availability monitoring on Wildberries, use the public API or parse product cards. The API is more stable but has strict limits. When parsing HTML, you must imitate real browser behavior: load images, execute JavaScript, save cookies between requests.

Ozon

Ozon uses less aggressive protection compared to Wildberries, but also requires a careful approach. The platform actively uses Cloudflare for bot protection, which adds an additional layer of complexity.

• Residential proxies are recommended, but quality datacenters can work with low request frequency
• Cloudflare challenge requires JavaScript execution — use headless browsers or special libraries
• IP rotation every 10-15 minutes is usually sufficient
• Delays between requests: 3-7 seconds
• Must save cookies after passing Cloudflare

Amazon

Amazon has a complex protection system that analyzes many request parameters. The platform is especially sensitive to request frequency and geographic location of IP addresses.

• Use residential proxies from the same country as the target marketplace (amazon.com — USA, amazon.de — Germany)
• Amazon often shows captcha with suspicious activity — prepare a system to solve it
• IP rotation every 15-20 minutes or every 50-100 requests
• Delays between requests: 5-10 seconds (Amazon is very sensitive to speed)
• Must use current User-Agent of popular browsers

Avito

Avito has a specific feature — geographic binding of listings. For correct stock availability monitoring in a specific city, you need proxies from that exact region.

• Use residential proxies with geolocation in the required city
• Avito shows different results for different regions — one IP can see only part of the listings
• Rotation every 10-15 minutes is sufficient
• Delays between requests: 3-5 seconds
• Platform uses cookies for session tracking — must save them

Request Limits and Delays Between Checks

Proper request frequency configuration is a balance between data relevance and block risk. Too frequent checks will lead to a ban, too rare — to loss of competitive advantage. Let's figure out how to find the optimal frequency for different scenarios.

Recommended delays between requests to one product:

Delays between sequential requests (when parsing multiple products in a row):

• Minimum delay: 2-3 seconds — absolute minimum for imitating human behavior. A real user cannot open product cards faster.
• Optimal delay: 4-7 seconds — recommended for most tasks. Add randomness: random(4000, 7000) milliseconds.
• Safe delay: 8-15 seconds — use for especially protected platforms or after receiving warnings about exceeding limits.

Important principle — randomness: Never use fixed intervals. Instead of exactly 5 seconds delay, make it random from 4 to 6 seconds. This is critically important for imitating human behavior. Real users don't click with perfect regularity.

Load distribution by time of day:
Marketplaces have peak load hours (usually evening 6:00 PM-10:00 PM) when protection systems are especially sensitive. It's recommended to reduce request frequency during these periods or pause monitoring altogether if data is not critical.

Ready-Made Tools for Monitoring Through Proxies

For stock availability monitoring, you don't necessarily need to write a parser from scratch. There are many ready-made solutions that support working through proxies and have built-in block protection mechanisms. Let's consider popular tools for different levels of technical preparation.

Ready-Made SaaS Services (No Programming)

1. Keepa (for Amazon)
Specialized service for Amazon monitoring. Tracks prices, stock availability, ratings. Has a built-in proxy system, requires no technical knowledge. Downside — works only with Amazon.

2. Parsehub
Visual parser that allows you to set up data collection through a graphical interface. Supports proxy configuration, IP rotation, delays between requests. Suitable for users without programming skills.

3. Octoparse
Parsehub alternative with more advanced automation capabilities. Has built-in templates for popular marketplaces, including Amazon, eBay. Supports cloud task execution and automatic proxy rotation.

Software Solutions (Require Configuration)

1. Scrapy (Python)
Powerful parsing framework for Python. Has built-in proxy support, rotation middleware, error handling. Requires programming skills but gives maximum configuration flexibility.

Example of proxy configuration in Scrapy is in the settings.py file — you add middleware for rotation and a list of proxy servers. The framework automatically distributes requests among available IPs.

2. Puppeteer / Playwright (JavaScript)
Headless browsers for Chrome/Firefox automation. Ideal for parsing sites with JavaScript and complex protection. Support proxy configuration at browser level, emulation of real user behavior.

3. Selenium
Classic browser automation tool. Works slower than Puppeteer but has more ready-made solutions and examples. Well suited for beginner developers.

Specialized Solutions for Marketplaces

1. MoySklad (integration with Wildberries, Ozon)
Accounting system with built-in integration with Russian marketplaces. Automatically synchronizes inventory, but works only through official APIs — not suitable for competitor monitoring.

2. Mpstats
Analytics service for Wildberries and Ozon. Collects data on products, prices, availability. Has its own proxy infrastructure, requires no additional configuration. Paid but stable.

3. SellerFox
Comprehensive solution for sellers on Russian marketplaces. Includes competitor monitoring, analytics, price automation. Works through its own proxies.

Step-by-Step Proxy Setup for Parser

Let's go through practical proxy setup for stock availability monitoring using a typical scenario: monitoring 100 products on Wildberries with checks every 30 minutes. The instructions will work for most ready-made parsers and custom solutions.

Step 1: Choosing and Purchasing Proxies

For Wildberries, you need residential proxies. Let's calculate the required quantity:
- 100 products × 2 checks per hour = 200 requests/hour
- Safe limit for Wildberries: 20 requests per IP
- Required: 200 ÷ 20 = 10 IP addresses minimum

It's recommended to get with reserve: 15-20 IPs for stable operation. When purchasing proxies, make sure the provider supports rotation by time or by requests.

Step 2: Getting Connection Data

After purchase, you'll receive data in the format:

Host: proxy.example.com
Port: 8080
Login: user123
Password: pass456
Type: HTTP/HTTPS or SOCKS5

Some providers provide a URL for automatic rotation:

http://user123:[email protected]:8080

This URL can be used directly in most parsers — the proxy server will change IPs according to settings.

Step 3: Configuration in Ready-Made Tool (Octoparse)

1. Open parsing task settings
In Octoparse, go to "Advanced Options" → "Proxy Settings"

2. Enter proxy data:

• Proxy Type: select HTTP or SOCKS5 (depending on your proxy)
• Server: proxy.example.com
• Port: 8080
• Username: user123
• Password: pass456

3. Configure rotation:
If your provider supports automatic rotation — simply enable "Use rotating proxy" option. If not — add IP list manually and enable "Rotate IP addresses" with 10-minute interval.

4. Configure delays:
In "Speed Settings" section, set delay between requests: 4-7 seconds (Random delay between 4000 and 7000 ms).

Step 4: Testing Settings

Before launching full monitoring, be sure to test settings:

• Check proxy connection — most tools have a "Test connection" button
• Run parsing of 5-10 products — make sure data is collected correctly
• Check logs for errors — codes 403, 429, 503 indicate problems with proxies or too high frequency
• Make sure IP is changing — different addresses should appear in logs according to rotation settings

Step 5: Monitoring and Optimization

After launch, regularly check:

• Success rate — percentage of successful requests should be above 95%
• Response codes — if 429 or 503 appear more than 5% — reduce load
• Proxy speed — if delays exceed 2-3 seconds, you might want to change provider
• Data relevance — compare obtained data with actual data on marketplace site

Gradually optimize settings: if everything works stably for a week, you can slightly increase check frequency or reduce delays. But do this gradually, step by step.

Common Mistakes and How to Avoid Them

Even with proper proxy configuration, beginners often make mistakes that lead to blocks or unstable parser operation. Let's examine the most common problems and ways to solve them.

Mistake 1: Using One User-Agent for All Requests

Problem: Parser sends all requests with the same User-Agent header, even when changing IP. This looks suspicious — the same "browser" appears from different addresses.

Solution: Rotate User-Agent together with IP addresses. Create a list of popular browsers (Chrome, Firefox, Safari of different versions) and randomly select from it with each proxy change. Most parsing libraries have built-in User-Agent rotation support.

Mistake 2: Too Frequent IP Rotation

Problem: Some configure IP change after each request, thinking it's maximally safe. Actually, this looks unnatural — real users don't change IP every second.

Solution: Use reasonable rotation intervals: 5-15 minutes or 20-100 requests per IP. This imitates the behavior of a real user who sits from one device for some time, then switches.

Mistake 3: Ignoring Cookies

Problem: Parser makes requests without saving cookies between them. Marketplaces use cookies for session tracking — if you don't save them, each request looks like a new session.

Solution: Always save cookies after the first request and pass them in subsequent ones. When changing IP, create a new session with new cookies — this imitates a new user.

Mistake 4: Fixed Delays Between Requests

Problem: A delay of exactly 5 seconds is set between all requests. This creates a perfect pattern that is easily detected by anti-bot systems.

Solution: Use random delays in a range. Instead of 5 seconds, make random(4, 7) seconds. Sometimes add longer pauses (10-20 seconds), imitating user distraction.

Mistake 5: Using Cheap Public Proxies

Problem: Trying to save on proxies by using free or very cheap public lists. These IPs are already blocked on most marketplaces, as they're used by thousands of other parsers.

Solution: Invest in quality residential proxies from reliable providers. This is the foundation of stable operation. Saving on proxies will lead to time loss fighting blocks and potential data loss.

Mistake 6: Lack of Error Handling

Problem: Parser doesn't handle errors and continues bombarding the server with requests even after receiving 429 (Too Many Requests) or 503 (Service Unavailable).

Solution: Implement smart error handling:

• When receiving 429 — double delays and change IP
• When receiving 503 — pause for 5-10 minutes
• When receiving 403 — immediately change IP and check settings
• Use exponential backoff: with repeated errors, increase pause exponentially

Conclusion

Stock availability monitoring through proxies is a complex but solvable task. The key to success is the right combination of quality proxies, proper rotation settings, and imitation of real user behavior. Start with conservative settings, gradually optimize them based on results, and always monitor logs for errors.

Remember that marketplaces constantly improve their protection systems. What works today may require adjustment tomorrow. Stay flexible, test new approaches, and don't be afraid to experiment with settings within reasonable limits.

Investing in quality proxies and proper configuration will pay off many times over through stable data collection, competitive advantage, and time saved on fighting blocks. Good luck with your monitoring!