Overview
In the competitive fast-moving consumer goods (FMCG) industry, efficient management of shelf and freezer space is crucial for maximizing product visibility and sales. Traditionally, planograms—detailed diagrams used to organize products on retail shelves—are manually implemented, monitored, and updated by brand representatives visiting stores. However, our client sought a more streamlined and automated solution to enhance control over product placement and improve retail shelf checks.
To meet this need, we developed a cutting-edge AI and computer vision-based solution, along with an intuitive dashboard, that empowers FMCGs to remotely monitor and manage product displays in stores with greater accuracy and efficiency.
The Challenge
FMCGs rely heavily on proper shelf placement to drive product visibility and sales, but the manual process of implementing and monitoring planograms is time-consuming and prone to human error. Ensuring that shelves are stocked according to planograms, identifying out-of-stock items, and tracking share of shelf can be difficult, especially across numerous retail locations. Our client needed an AI-driven solution capable of detecting product placement and providing actionable insights through image recognition, without requiring frequent in-person visits from brand representatives.
The Solution
We developed an advanced image recognition system integrated with a user-friendly dashboard to automate and optimize the management of retail shelf space. This AI-powered solution can analyze shelf images, detect SKUs, track product placement, and assess planogram compliance, providing FMCGs with critical insights into shelf performance. Key features of the solution include:
SKU Detection & Shelf Mapping
The AI system is trained on sample images to detect and identify SKUs on shelves, even in complex scenarios such as adjacent shelves, non-uniform rows, or side-view images. It accurately maps shelf boundaries and product rows to provide a clear visual representation of product placement.
Share of Shelf & On-Shelf Availability Estimation
The system automatically estimates key metrics such as Share on Shelf, On-Shelf Availability, and Item Count. These insights help FMCGs understand how much shelf space their products occupy and identify potential out-of-stock situations.
Planogram Compliance Recognition
The AI compares the real-time shelf display with the required planogram to ensure compliance. It highlights discrepancies and provides recommendations for corrections, enabling efficient shelf management and reducing the need for manual checks.
CTA Banner Detection
In addition to product recognition, the system detects and reads Call to Action (CTA) banners displayed on shelves, providing a comprehensive view of the entire retail setup.
Tech Stack
Our development team utilized advanced AI and image recognition technologies to ensure that the solution could process and analyze shelf images at a scale. The key technologies involved in building this solution include:
- AI-Based Image Recognition for SKU detection and shelf mapping.
- Python for data processing and machine learning model development.
- Cloud Infrastructure for scalable storage and real-time analysis.
- Custom Dashboard for visualizing insights and planogram compliance.
Conclusion
By leveraging AI and computer vision, our solution revolutionizes how FMCGs manage retail shelf space. The automated system provides real-time insights into product placement, stock levels, and planogram compliance, allowing brands to optimize their retail displays without the need for frequent in-person checks. FMCGs now have greater control over their shelf presence, ensuring that their products are always displayed optimally to maximize visibility and sales.
This AI-powered solution is a game-changer for retail operations, offering an efficient and scalable way to manage shelf space across multiple locations, while reducing operational costs and improving compliance.
