A data-driven analysis examining the location strategies of coffee franchises in South Korea, specifically investigating the proximity relationship between Starbucks and its competitors.
This project investigates the business hypothesis popularized by Ediya Coffee's strategy of "defending the seat next to Starbucks"—the idea that positioning competitor coffee shops near Starbucks locations can be beneficial for business due to overflow traffic and price differentiation.
Historically, the strategy was based on the premise that:
- During peak hours (like lunchtime), Starbucks stores often reach capacity.
- Customers unable to find seats would naturally migrate to nearby alternatives.
- Price-conscious consumers would prefer lower-priced alternatives when available nearby.
- Interestingly, Starbucks locations near competitors often showed higher sales than isolated locations.
This project aims to verify with data whether this proximity strategy is widely adopted across the Korean coffee market.
Project Duration: June 30, 2025 - July 2, 2025
- Analyze the spatial distribution of Starbucks locations across Seoul, South Korea.
- Examine the proximity patterns between Starbucks and other major coffee franchise locations.
- Validate the hypothesis that competitor coffee shops strategically locate near Starbucks stores.
- Provide data-driven insights for coffee franchise location planning.
-
Conclusion: The Proximity Strategy is Demonstrably True.
- Low- to mid-priced coffee franchises show a clear tendency to open stores in very close proximity to Starbucks, aiming to leverage the foot traffic generated by the established brand.
-
The Optimal Distance is "30m to 50m."
- Competitor stores were most frequently found within a "hyper-proximal" range of 30 to 50 meters. This suggests an intentional strategy to directly attract the attention of customers heading to Starbucks, competing primarily on price.
-
The Most Aggressive Franchise is "Mega Coffee."
- Across all analyzed distances (30m, 50m, 100m), Mega Coffee had the highest number of stores located near Starbucks, making it the volume leader in this strategy.
-
The Most Consistent Franchise is "Paik's Coffee."
- Paik's Coffee had the lowest average distance to Starbucks across all its locations (approx. 306m). This indicates that its overall brand strategy is the most consistently aligned with shadowing Starbucks.
-
The Analysis is Focused on "Urban Areas."
- This analysis is based on data from Seoul (e.g., Gangnam-gu, Jongno-gu) and therefore reflects store location patterns in a dense, urban environment.
The data shows that as the distance to Starbucks decreases, the concentration of specific brands increases.
- Within 100m: This is the range where the proximity strategy becomes clearly visible. The average distance for stores already within this radius is actually much closer, between 56m and 66m.
- Within 50m: Competition intensifies. Mega Coffee (28.28%), Paik's Coffee (18.62%), and Compose Coffee (18.62%) hold significant shares.
- Within 30m: Mega Coffee (30.3%) and Compose Coffee (19.7%) are the most aggressive in employing a hyper-proximity strategy.
Conclusion: The "30m to 50m" range appears to be the optimal sweet spot for maximizing competitive impact.
This analysis clearly reflects an "Urban" pattern.
The location data is from districts like Gangnam-gu and Jongno-gu in Seoul, which are among the most commercially dense and high-traffic areas in South Korea. The proximity strategy is highly effective in such environments, targeting customers on foot.
It is expected that these patterns would differ in suburban or rural areas, where factors like parking availability and locations within large shopping malls play a more significant role. (Note: This analysis is limited to the urban data provided).
Different brands employ this strategy with varying approaches.
| Rank | Criteria | Brand | Description |
|---|---|---|---|
| 1 | Volume | Mega Coffee | Operates 143 stores within a 100m radius of a Starbucks. The most aggressive, volume-based strategy. |
| 2 | Consistency | Paik's Coffee | Has the lowest overall average distance (306.5m). Its brand-level strategy is the most consistently tied to Starbucks locations. |
| 3 | Hyper-Proximity | Compose Coffee | Holds the second-highest share (19.7%) within the hyper-proximal 30m range, trailing only Mega Coffee. |
| - | Differentiation | Ediya Coffee | Despite a large number of stores, its average distance is relatively far (389.4m), suggesting a strategy focused on different niches (e.g., residential areas, specific office buildings). |
Low- to mid-priced coffee franchises in Seoul successfully employ a "parasitic" or "symbiotic" location strategy that leverages Starbucks's powerful brand and market research capabilities. The data clearly shows that they are "free-riding" on the A-grade commercial zones identified by Starbucks, effectively attracting price-sensitive consumers at a fraction of the market analysis cost.
- Web Scraping: Use Selenium to crawl Starbucks location data from the official website.
- Data Processing: Convert scraped data to structured DataFrames using pandas.
- External Data: Download and integrate CSV files containing other major coffee franchise locations in Korea.
- Distance calculations between competing franchises using the Haversine formula.
- Visualization of spatial relationships in Jupyter Notebooks.
- Python - Primary programming language
- Selenium - Web scraping automation
- Pandas - Data manipulation and analysis
- CSV - Data storage and exchange format
- Jupyter Notebooks - Analysis and documentation
- Clone the repository
git clone [https://github.com/yourusername/coffee-location-strategy.git](https://github.com/yourusername/coffee-location-strategy.git)
cd coffee-location-strategy- Install dependencies
pip install -r requirements.txt- Run the data collection script
python DataAnalysis_Starbucks\starbucks.py
This script generates a dataset of Starbucks locations in Seoul, South Korea (Store name, Address, Latitude, Longitude).
6._Location_strategy_for_a_coffee_franchise.ipynb contains the full data analysis process, documentation, and visualization.
- Please open in colab to see visualization (graph, tables)
- Starbucks Korea official website (https://www.starbucks.co.kr/index.do)
- Public datasets of major Korean coffee franchises (extracted from the Small Business Market Promotion Agency Shopping District Information dataset)
This project is licensed under the MIT License.
For questions or collaboration opportunities, please open an issue or contact the project maintainer.