A dark store is an order fulfillment warehouse closed to regular shoppers. It may look like a supermarket from the outside, but inside there are no shelves for browsing, no checkouts, no customers — only racks of products, pickers, and couriers. The model exists for one purpose: cutting delivery time to 15–30 minutes by placing the warehouse close to the buyer.

For companies in e-grocery and food-tech, competitors' dark stores have become a critical market variable. Each location has its own assortment, its own prices, and its own delivery terms. Understanding what a competitor offers in a specific neighborhood is impossible through traditional methods — data changes too fast. The only reliable source is the competitor's mobile app in real time.

How a Dark Store Works

A dark store solves a logistics problem a regular store can't: delivering an order in 15 minutes. To do this, the warehouse is placed not on the city outskirts but in residential neighborhoods — where customers actually live. A typical dark store occupies 300–800 sq. m. and carries 1,500 to 5,000 of the most in-demand SKUs.

The key difference from a conventional distribution center is speed. An order is picked in 2–5 minutes, then a cyclist or scooter courier delivers it within a 1–2 km radius. That's why you need many dark stores, densely distributed: the denser the network, the wider the coverage and the faster the delivery.

This model differs fundamentally from express delivery out of a regular supermarket. A dark store is optimized exclusively for picking — it has no retail function. These are different operational models with different speeds and economics.

The Quick-Commerce Market: Key Players and Scale

Russia is one of the most developed quick-commerce markets in the world. As of late 2025, two segment leaders manage thousands of locations nationwide:

Samokat

Over 2,150 dark stores across 129 Russian cities. Delivery in 15–30 minutes. The 2,000th dark store opened in May 2024; the network added ~150 new locations in the first 9 months of 2025.

Yandex Lavka

450+ dark stores, delivery in 10–15 minutes. Assortment of 2,500+ SKUs per location. Added 91 new locations in Q1–Q3 2025. Actively developing a franchise model for regional expansion.

Beyond the two leaders, the segment includes VkusVill (147+ dark stores), Sbermarket Express, Kuper, and regional players. The global dark store market is projected to reach $177 billion by 2031, growing at a CAGR of 36.9% — Russia remains one of the flagship markets for this model. Such density of competing services means prices and assortments shift constantly, making systematic competitor monitoring a competitive necessity.

What You Can Learn from a Competitor's App

A quick-delivery service's mobile app is the primary source of competitor intelligence. It's how customers see current assortment, prices, and stock availability at their specific address. The same data is available for systematic analysis.

Prices per Dark Store

App prices are geo-localized. The same product can be priced differently across different neighborhoods in the same city.

Assortment per Location

Which SKUs exist at a specific dark store, which ones the competitor added or removed — tracked per location.

Inventory Levels

Actual stock at a location in real time. When a competitor goes out of stock — it's visible immediately.

Promotions and Offers

Promotional campaigns, discounts, personalized offers — what competitors launch and in which cities.

Coverage Geography

Which neighborhoods a competitor serves, where their coverage gaps are — opportunities for your own expansion.

Change Dynamics

How prices and assortments shift over time: response to holidays, seasonality, new SKU launches.

Use Cases: Competitor Analysis in E-Grocery

Data from competitor apps solves concrete operational problems — not just for delivery services themselves, but for the suppliers and manufacturers who work with them.

Price monitoring across all locations. A manufacturer wants to know at what prices their products are listed across competing delivery services in different neighborhoods. Apps like Samokat and Yandex Lavka return geo-localized prices: the same SKU can be checked across dozens of cities simultaneously. Without mobile app scraping, such coverage is unachievable. Data is refreshed on a schedule — hourly or daily — and delivered in a format ready for BI integration or spreadsheet analysis.

Competitor out-of-stock monitoring. When a competing delivery service runs out of stock in a popular category, it's a window of opportunity: boost advertising, temporarily lower prices, attract customers who couldn't find the item at a competitor. Detecting the out-of-stock moment in real time is a job for automated app scraping. This type of monitoring is especially valuable for manufacturers in short shelf-life categories — dairy, snacks, ready meals — where reaction speed directly affects revenue.

Assortment analysis. A supplier wants to understand which of their products are listed at competing retailers and which aren't. Comparing assortments across 2,000+ Samokat dark stores or 450+ Yandex Lavka locations gives a complete picture of national distribution — without a single call to a sales representative.

Who Needs E-Grocery Competitor Analysis

  • Quick-delivery services — for monitoring competitor prices and assortment in every city and neighborhood they operate in.
  • FMCG manufacturers and brands — to see shelf prices of their products and competitors' across different delivery services.
  • Suppliers and distributors — for dark store distribution analysis and identifying gaps in market coverage.
  • Investors and market analysts — to assess growth dynamics and pricing strategies of e-grocery players.
  • Product teams — to understand how competitors structure their assortment and respond to seasonal demand.

How the Process Works

1. Set Geolocations

You specify cities, neighborhoods, or exact addresses. The system simulates requests from each location — exactly as a customer would in the app.

2. Collect Data

From competitor apps, we extract prices, stock levels, assortment, and promotions — for each dark store at the specified geolocations.

3. Deliver Results

Structured data in JSON, CSV, or Excel. One-time or on a schedule — daily, multiple times per day, as required.

Need competitor intelligence from e-grocery apps?

Tell us about your task — we'll find the right monitoring format. Samokat, Yandex Lavka, and other delivery services, any cities and neighborhoods, at the frequency you need.

Discuss your task

Three Monitoring Scenarios for E-Grocery Competitors

Scenario 1: Reacting to competitor promotions in real time. An FMCG manufacturer wants to know when Samokat launches a discount on a rival brand in a specific city. Scraping the app every hour sends an alert the moment the promotion starts — giving several hours to respond: adjust pricing, boost advertising spend, or redirect trade marketing budgets across channels.

Scenario 2: Identifying distribution gaps. A beverage supplier wants to know in which cities its products are absent from Yandex Lavka while competitors' analogues are already listed there. Comparing assortments across 450+ locations overnight delivers a complete gap map — without calling account managers or going through lengthy purchasing negotiations.

Scenario 3: Pre-season price monitoring. A week before New Year's or a major holiday, competitors start raising prices on seasonal categories: chocolate, alcohol, gift sets. Daily monitoring lets you track the exact moment prices change at each dark store and make a data-driven decision — hold price and capture traffic, or follow the market upward.

Quick commerce is a market where advantage is measured in city blocks and hours of reaction time. Without systematic monitoring of competitor apps, pricing and assortment decisions are made blind. Mobile app scraping lets you see the market the way customers see it: in real time, down to a specific dark store in a specific neighborhood.

Samokat has over 2,150 dark stores — that's 2,150 separate price points, 2,150 assortment matrices, and 2,150 locations to track. The only way to do this systematically is mobile app scraping.