Telegram
Email
Wildberries, Ozon, and Avito are the leading platforms of the Russian-language internet for trading and analytics. Sellers and analysts collect prices, stock levels, reviews, and listing rankings from them. But when you gather data from a single IP, the platforms quickly kick in rate limits and CAPTCHAs. Let's break down which proxies you need to scrape marketplaces and how to set up rotation so you don't get banned.
Why scrape marketplaces
- Monitoring competitors' prices and automatically adjusting your own.
- Collecting reviews for analytics and SEO listing descriptions.
- Scraping the catalog and categories for analytics dashboards.
- Tracking stock levels and search rankings.
Why you need proxies
Marketplaces limit the number of requests from a single IP. Exceed the limit and you'll get a CAPTCHA, while the address is temporarily blocked. Rough guidelines for the limits (they vary by platform and may change):
| Platform | Approximate limits | Proxy type |
|---|---|---|
| Wildberries | ~60–120 requests/min per IP, then a CAPTCHA | Russian server/residential with rotation |
| Ozon | Aggressive anti-bot, behavioral checks | Residential/mobile with rotation |
| Avito | Fast datacenter ban, CAPTCHA | Russian mobile/residential |
Ready-made landing pages with pricing: proxies for Wildberries, proxies for Ozon, proxies for Avito.
Which proxies to choose
- Scraping the catalog and prices — Russian server or residential proxies with a large pool and rotation.
- Working with the seller dashboard — a static mobile IP so the session is treated as "live".
- Geo — Russia. Marketplaces serve data differently to users from Russia and from abroad.
How to set up rotation for scraping
- Get a pool of IPs or a mobile proxy with automatic switching.
- Distribute your requests: no more than the platform's limit per IP per minute.
- Add delays between requests and CAPTCHA handling.
- Use SOCKS5/HTTP — both work for scrapers (see our comparison of the protocols).
Our proxies work correctly with popular tools: Scrapy, Python requests, A-Parser, Netpeak Checker. The general principles of data collection are on our proxies for scraping page.
Case study: price monitoring on Wildberries
An analyst was collecting prices across 100,000 listings from a single IP and hit a CAPTCHA every half hour. The solution was a pool of IPs with rotation: requests are distributed so that each address stays within the per-minute limit. After that, data collection ran without interruptions. A separate static mobile IP was assigned for the supplier dashboard.
FAQ
How many requests can one IP handle on WB?
Roughly 60–120 requests per minute without a CAPTCHA; exceed that and the IP is temporarily blocked. That's why large volumes require rotation.
Do I need rotation for scraping?
For volumes from tens of thousands of listings per day — yes, definitely, otherwise you'll trip the anti-bot system.
Can I run a seller account and scrape from the same proxy?
No. The seller account needs a stable mobile IP; scraping needs a separate pool with rotation.
Where can I get proxies for marketplaces?
On our WB, Ozon, and Avito pages, or in the general PROXYLEET catalog.
