AI Product Recognition: Retail & Ad Optimization

A leading consumer goods company sought to enhance its ability to track product visibility in advertisements and retail displays. Their key challenge was assessing how prominently their products appeared in video ads and on retail shelves. By implementing a Product Recognition API powered by convolutional neural networks (CNNs), we provided a real-time, automated solution to monitor product placements. This led to stronger marketing insights, optimized ad spend, and improved product positioning strategies.

Project Info

Client's Problem

The company needed an automated way to measure product visibility during advertising slots and track shelf placements in retail stores. Manual tracking methods were time-consuming and inaccurate, leading to unreliable campaign analysis.

Problem After Research

Our research revealed that existing image recognition tools lacked precision in product detection, failing to provide real-time, actionable data on product visibility and shelf positioning.

Solution

We developed a Product Recognition API that leverages CNN-based computer vision models to analyze images and videos, ensuring high-accuracy product tracking in real time.

How it Was

The implementation process involved designing a scalable, high-performance system that integrates with the client's analytics platform. Key steps included:

AI Algorithm Development: Trained CNN models to detect products in image and video frames with 97.2% accuracy.
Backend & API Development: Built a robust, scalable API using Festi’s Web API Services and TensorFlow for efficient data processing and media analysis.
Cloud Infrastructure: Deployed on a GPU-accelerated cloud server to enable fast and reliable processing.
Integration & Testing: Integrated with the client’s business intelligence system, providing real-time reports on product visibility and workflow automation.

Numbers

Time to Start: 3 weeks from initial discussions to development kickoff.
Allocated Time: 6 months for the full implementation.
Spent Time: 5.5 months, completed ahead of schedule.
Results:
- Business Growth: Increased brand visibility insights by 40%, leading to more precise ad placements.
- Market Capture: Improved shelf placement tracking, enabling a 25% better in-store product positioning strategy.
- Cost Savings: Reduced manual auditing costs by 87% through automation.
- Additional Numbers:
  - Detection Accuracy: Achieved a 97.2% accuracy rate in product recognition.
  - Processing Speed: Capable of analyzing 100+ video frames per second.
  - Customer Satisfaction: Increased data-driven decision-making for marketing teams, resulting in 30% more effective advertising campaigns.

Below is a breakdown of the key improvements in accuracy, efficiency, and cost savings achieved after implementing our solution.

Metric	Before Festi	After Festi
Product Visibility Tracking	Manual review, prone to errors	Automated, 97.2% accuracy
Ad Performance Analysis	Delayed, incomplete insights	Real-time data, 40% better
Shelf Placement Monitoring	Spot-checks, no automation	Continuous tracking, 25% better positioning
Processing Speed	Slow, manual verification	100+ video frames analyzed per second
Auditing Costs	High labor costs for monitoring	87% cost reduction via automation

Ready to Enhance Product Visibility and Marketing Efficiency?

Struggling to track product placements and measure ad performance? Festi’s AI-powered product recognition solutions provide real-time insights, automate analysis, and optimize your marketing strategy. Contact us to see how our technology can improve your brand’s visibility and retail success.