A seller sets up price monitoring, sees beautiful graphs in the table — and decides to lower the price on a product that is actually already cheaper than all competitors. Familiar situation? The problem is not with the idea of monitoring itself, but with how the data is collected. We analyze seven mistakes that turn the price control system into a source of false information and show how to fix it in practice.
Why Price Monitoring Accuracy is Critical for Business
Monitoring competitor prices on Wildberries, Ozon, Avito, and Yandex.Market is not a one-time task but an ongoing process that directly affects profitability. If data is collected with errors, the seller either engages in unnecessary price dumping or misses the opportunity to raise prices where competitors are more expensive. In a catalog of 500-1000 SKUs, even 5-10% of inaccurate data can translate into thousands of rubles in lost profit each month.
The problem is that marketplaces actively protect against automated data collection: they show different prices depending on the region, device, order history, and also block suspicious activity with CAPTCHAs and temporary IP bans. If the monitoring system does not account for these mechanisms, it collects not real market prices but a distorted picture — and the business makes decisions based on fake data.
Below is an analysis of specific mistakes that occur most frequently, with explanations of why they arise and how to eliminate them without involving developers.
Mistake 1: Data Collection Without IP Rotation — Blocks and CAPTCHAs
The most common mistake is launching monitoring from a single static IP address or from a data center server without rotation. Wildberries and Ozon see hundreds of requests from one IP in a short period of time and either start showing CAPTCHAs, providing distorted data (for example, "product unavailable" or outdated price from cache), or completely block access.
As a result, the monitoring system either does not receive data at all or receives it partially — and gaps appear in the report, which many interpret as "the competitor does not have this product," when in fact it is just a block from the platform.
The solution is to use a pool of IP addresses with automatic rotation for each request or at a specified interval. For price monitoring tasks on marketplaces, residential proxies are well suited: they use real IP addresses of ordinary internet users, so they appear to the platform as organic traffic rather than a bot network. This reduces the frequency of CAPTCHAs and blocks by tens of times compared to data center addresses without rotation.
| Type of Proxy | Suitable For | Risk of Blocking |
|---|---|---|
| Data Center Proxy | Fast collection of small catalogs without strict protection | High on protected platforms |
| Residential Proxies | Regular monitoring of Wildberries, Ozon, Avito | Low |
| Mobile Proxies | Checking mobile prices and promotions in apps | Minimal |
Mistake 2: Ignoring Geolocation and Regional Prices
Wildberries and Ozon show different prices depending on the shipping warehouse, delivery region, and even specific city. A product may cost 1200 rubles for a buyer from Moscow and 1450 rubles for a buyer from Vladivostok — due to different logistics and availability at regional warehouses.
If monitoring is launched from a single IP tied to one region, you only get the price for that region and mistakenly take it as the "competitor's price" in general. This is especially critical for sellers who sell in multiple regions of Russia or work with different warehouses of the marketplace.
The correct approach is to collect prices from several geographic points, simulating buyers from different cities. For this, proxies with geo-targeting for specific regions of Russia are needed. Residential proxies with the ability to select a city or region allow building a complete price map across the country, rather than being limited to one point. This is especially important for products with significant differences in logistics costs — clothing, large household appliances, furniture.
Mistake 3: Incorrect Collection Frequency — Outdated Data
Many set up price monitoring once a day or even once every few days, thinking that this is sufficient. But competitors on Wildberries and Ozon can change prices several times a day — especially during sales, "Product of the Day" promotions, or flash discounts that last only a few hours.
If your system collects data once a day, you either miss short-term promotions from competitors (and lose sales at that moment), or conversely — react to a price that has long since changed back, and engage in unnecessary dumping.
The optimal frequency depends on the product category: for highly competitive niches (electronics, cosmetics, children's products), collection every 2-4 hours is recommended, while for less dynamic categories, 1-2 times a day is sufficient. As the frequency of collection increases, so does the load on the infrastructure — this is where IP rotation through residential proxies becomes mandatory; otherwise, frequent requests from the same addresses will quickly lead to blocking.
Mistake 4: Lack of Real User Emulation
Marketplaces analyze not only the IP address but also behavioral patterns: the speed of transitions between pages, the presence of browser headers, cookies, user-agent, cursor movements. If requests are made "head-on" without emulating a real browser, the platform easily distinguishes a bot from a human and shows protective pages or distorted content.
For sellers who do not engage in coding, the solution is to use ready-made anti-detect browsers: Dolphin Anty, AdsPower, Multilogin, Octo Browser. These tools allow you to create profiles with unique digital fingerprints and associate a separate proxy address with each profile. Thus, each "virtual buyer" who visits Wildberries to check the price appears as a unique real person, not as part of a bot network.
The combination of an anti-detect browser + residential or mobile proxy is a working scheme used not only by arbitrageurs for farming advertising accounts but also by sellers to build a reliable price monitoring system without constant blocks.
Mistake 5: Ignoring Personalization and A/B Pricing
Marketplaces are increasingly using price personalization: the same product can be shown at different prices depending on search history, authorization in the personal account, participation in loyalty programs (for example, Wildberries Wallet), or even random A/B pricing tests.
If monitoring is launched from an authorized account or from a "warmed-up" profile with a purchase history, you may receive a personalized discounted price that does not reflect the real market situation for a new buyer. Conversely, if a competitor sets hidden promotions only for subscribers, regular anonymous parsing will not see them.
To obtain the most objective picture, it is recommended to combine two collection modes: anonymous (without authorization, clean profile) for the basic market price and authorized (with a test account) for tracking personalized offers and loyalty promotions. Both modes should use different, non-overlapping IP pools so that the platform does not link them into one session.
Mistake 6: Issues with Dynamic Content and JS Rendering
Product cards on Wildberries and Ozon heavily rely on JavaScript: prices, stock, discounts are loaded dynamically after the initial page load. If the monitoring tool only receives the raw HTML without executing scripts, it often sees empty fields or outdated prices captured in the page cache before applying dynamic discounts.
This is especially noticeable during promotions like "price when added to cart" or "discount with promo code," where the final price is formed only after certain actions on the page. A simple request without full page rendering will not see such a price and will capture an incorrect value.
For ready-made solutions (without coding), this problem is usually solved by specialized marketplace parsing services that already account for dynamic content loading. When choosing such a service, be sure to check if it mentions support for JS rendering and current prices "considering promotional discounts," not just the base price from the card.
Mistake 7: Lack of Validation of Collected Data
Even with properly configured data collection, errors are inevitable: network failures, temporary blocks, changes in the marketplace page structure. If the monitoring system lacks a validation step for the collected values, anomalous data goes directly into the report and affects pricing decisions.
A classic example: a product that cost 2000 rubles suddenly "dropped" to 20 rubles in the report — this is almost always a parsing error (for example, the price per unit of measurement was captured instead of the package price), not a real competitor sale. Without automatic checks for anomalous deviations, such errors can easily be mistaken for real dumping and lead to a price war that is unnecessary.
A simple validation rule: if the new price differs from the previously recorded value by more than 50% in either direction, the system should mark the record as "requires verification" and not automatically pass it to the pricing decision module. This is a basic filter that eliminates most gross data collection errors.
Checklist for Proper Price Monitoring
Before launching or reviewing a competitor price monitoring system, go through the following points:
- IP rotation is used through residential or mobile proxies, not a static data center address
- Data collection occurs from several regions relevant to your sales geography
- Collection frequency matches the dynamics of the product category (from 2 hours to once a day)
- Requests emulate a real browser (through an anti-detect browser or service with fingerprint support)
- There is a separation between anonymous and authorized data collection to account for personalization
- The tool supports JS rendering to obtain the final price considering discounts
- Automatic validation of anomalous price deviations is set up before passing to the report
- Data is stored with a history of changes, not just the current value — this helps see competitor patterns
| Mistake | Consequence | Solution |
|---|---|---|
| Without IP Rotation | CAPTCHAs, blocks, data gaps | Residential proxies with automatic rotation |
| Ignoring Geolocation | Incorrect price for another region | Proxies with geo-targeting for Russian cities |
| Rare Collection | Missing short-term promotions | Increase collection frequency to 2-4 hours |
| No Browser Emulation | Protective pages instead of data | Anti-detect browser + proxy for each profile |
| No Validation | Anomalous prices in reports | Automatic deviation checks |
Conclusion
Monitoring competitor prices on Wildberries, Ozon, Avito, and other marketplaces brings real benefits only when data is collected accurately and regularly. The seven mistakes described above — blocks due to static IP, ignoring geolocation, incorrect collection frequency, lack of browser emulation, price personalization, issues with dynamic content, and lack of validation — are encountered by almost every seller at the start, but all of them can be eliminated without involving programmers.
If you are just setting up a price monitoring system or noticing that the current data looks suspiciously stable or, conversely, too chaotic, start by checking the collection infrastructure. For regular monitoring of catalogs on marketplaces, we recommend trying residential proxies — they minimize the risk of blocks and allow data collection from different regions as if real buyers were doing it. And for checking mobile versions of applications and promotions available only in mobile traffic, consider mobile proxies — they provide an additional level of data reliability in scenarios where the desktop version shows a different picture.