API Support

In today’s technology-driven landscape, shelf recognition systems are rapidly transforming how companies operate. According to Allied Market Research, the global image recognition market was valued at $28.3 billion in 2022 and is projected to reach $126.8 billion by 2032 — growing at a CAGR of 16.5% from 2023 to 2032.

This impressive growth is driven largely by the rising demand for automation, as FMCG businesses strive to improve efficiency and accuracy across retail execution. At the same time, innovations in machine learning and computer vision continue to redefine what’s possible in shelf recognition and retail analytics.

The Evolution of Shelf Recognition Technology

The development of shelf recognition solutions began around 2015, following the Gartner Hype Cycle for Computer Vision. During the early “technology trigger” phase, expectations were sky-high — many believed that image recognition would instantly solve all operational challenges. By 2017, investment and enthusiasm peaked, with businesses anticipating seamless and error-free recognition systems.

At that stage, recognition accuracy and processing speed were the main benchmarks for success. However, by 2021, the market entered a trough of disillusionment as companies realized that identification alone wasn’t enough. The focus shifted from basic detection to real-time data analytics, actionable insights, and smarter methods of integrating shelf recognition into daily workflows.

By 2023, shelf recognition technology had matured and approached the “plateau of productivity.” No longer just a support tool for field teams, it became a critical source of business intelligence. The addition of technologies like Augmented Reality made shelf scanning faster and more interactive, allowing teams to capture shelf data in seconds while instantly visualizing results.

Manual Shelf Recognition: Limitations of Traditional Audits

For decades, FMCG companies relied on manual shelf recognition to audit retail shelves. This method involved employees physically inspecting shelves to verify product placement, confirm assortment, and ensure planogram compliance according to FMCG merchandising standards.

While this approach provided some visibility, it came with serious drawbacks: it was time-consuming, expensive, and highly dependent on human judgment. The process was also vulnerable to bias and inconsistency, leading to unreliable retail audit data and missed performance insights.

For example, one of our UK-based clients used manual audits to assess field sales performance through KPIs like shelf facings per SKU. According to company policy, an SKU counted toward KPI achievement only if displayed without any extra packaging. While the team met these targets consistently under manual control, the introduction of AI-based shelf recognition revealed a different reality.

When automation replaced manual counting, the system detected that many SKUs were displayed incorrectly — often with packaging still attached. When questioned, team members gave conflicting explanations about what qualified as a compliant display, exposing the subjectivity of manual processes.

This misalignment ultimately caused over 50% of KPIs to fail, leading to lost sales opportunities and reduced revenue. The example highlights a crucial lesson: without reliable shelf recognition powered by AI, even well-defined performance metrics can become distorted by human error.