How to Choose the Best Retail Data Capture Methods for Shelf Intelligence


Shelf intelligence solutions are no longer on the shopping wish list. They are a modern retail competitive necessity.

In short

  • This guide compares four retail shelf data capture methods: mobile devices, robots, fixed cameras, and wireless cameras.
  • Learn from over 20 engagements and deployments with leading retailers in the USA and Europe, considering store size, update frequency, and ease of deployment.
  • See how to choose and combine data capture methods based on cost, strategic, and operational factors. The right approach can reduce costs, ensuring smoother rollout, scaling, and operations.
  • Explore how a hybrid approach combining multiple capture methods often provides the best balance of cost, coverage, and scalability, resulting in higher ROI.

Shelf intelligence solutions are a modern retail competitive necessity. They turn shelf images into actionable insights - like out-of-stocks, price errors, and planogram issues - so retailers can fix problems faster and boost store performance.

Shelf Intelligence is crucial for retailers to maintain optimal stock levels, improve sales, and enhance customer satisfaction.

These solutions typically involve three core components:

  • Hardware: for retail data capture.
  • AI-powered insights engine: for processing data and generating actionable insights for store associates.
  • Reporting: for understanding store performance, providing comprehensive dashboards for store, regional, and category management.

Historically, retailers have approached shelf intelligence with understandable caution. The primary hesitation often stemmed from viewing the retail data capture component as a significant upfront hurdle, with substantial implementation costs and lengthy timelines.

It often felt like the hardware was simply an expense, with the actual benefit - the insights driving actions to resolve critical shelf issues - seen as a separate, distant outcome.

However, choosing the right approach to retail data capture hardware can help retailers mitigate cost concerns and ensure smooth rollout and scaling of solutions, while also unlocking additional operational benefits.

How to choose the right retail data capture method

This guide examines various methods for capturing retail shelf data to enhance on-shelf availability, inventory accuracy, and overall shelf health. It’s based on over 20 engagements and deployments with leading retailers in the USA and Europe, taking into account factors such as store size, update frequency, and ease of deployment.

Our analysis focuses on four core retail data capture techniques available in the market: mobile devices, autonomous robots, fixed cameras, and wireless cameras. It’s important to note that using these methods in isolation has limitations. A mixed or hybrid approach often yields the best results. We explore this in detail below.

A table exploring core retail data capture methods

Mobile Devices

Autonomous Robots

Fixed Cameras

Wireless Cameras

Store associates use mobile devices, like smartphones and handheld computers to capture shelf data with specialized applications.

Robots that slowly roam the store, capturing data on inventory, pricing, and product locations.

Fixed cameras mounted on walls, ceilings, or elevated fixed points monitor specific areas of the store regularly.

Battery-powered cameras, mounted on shelves or specific areas, monitor certain areas of the store regularly.

Key points:

  • Cost-effective use of existing hardware
  • Flexible and easy to implement
  • Easily scalable

Key points:

  • Low labor requirements
  • Speed and ease of deployment
  • High innovation appeal

Key points:

  • Low labor requirements
  • High-frequency data capture
  • No need for battery maintenance or replacement

Key points:

  • Low labor requirements
  • High-frequency data capture
  • Cableless cameras, ease of deployment

Cost, operational, and strategic factors

When evaluating retail data capture methods, retailers should consider both cost factors (such as device and implementation costs) and operational and strategic factors (such as accuracy and employee acceptance) to help determine the best route.

Cost analysis

The economics of retail data capture methods depend on the store size (number of SKUs), and the frequency of data capture required (updates per week). A good estimate for the required updates per week is the number of deliveries to the store per week.

Our analysis is based on the cost to cover 1000 products for a whole year [$/1000 products per year]. According to this metric, the following chart shows which capture method is most cost-effective in specific scenarios. While fixed cameras and wireless cameras have different cost structures, their total cost of ownership is generally similar.

Specific capture methods are most cost-effective for certain store sizes and update frequencies. It can be summarized using the following five areas:

  1. Mobile device data capture is the most cost-effective* option for smaller convenience and city store formats with lower update frequency (1 to 5 updates per week).
  2. Robots are the cheapest option for supermarkets and hypermarkets, which manage over 40,000 SKUs and require daily updates.
  3. Medium to large-sized supermarkets that require daily updates present an interesting area where cost factors intersect. Operational and strategic factors become crucial decision-makers (see below).
  4. Mobile devices and fixed/wirelesscameras are most feasible in medium to small city and convenience stores with daily capture needs.
  5. Fixed/wireless cameras are the most cost-effective option for stores that require multiple daily or hourly updates.

* Mobile data capture costs depend on average hourly labor costs in the USA. Lower labor costs expand the orange (mobile) area, while higher costs reduce it.

Operational and strategic factors analysis

While cost is a primary consideration, successful shelf intelligence deployments also hinge on a range of operational and strategic factors. These elements dictate how seamlessly a solution integrates into daily store operations, its impact on associates and customers, and its overall strategic value.

The operational and strategic analysis considers many factors; however, for our research, we focused on nine and scored them on a scale of 1 (very poor) to 5 (very good).

As this matrix illustrates, each data capture method offers a unique profile of strengths and considerations. Understanding these nuances is vital for selecting the optimal solution:

  • Mobile capture stands out as the most flexible and easiest method to deploy as it leverages devices already in use, with minimal impact on customers. It also boasts strong connectivity due to its inherent flexibility and offers remarkable store format support by adapting to any store layout and any shelf. Compared to other methods, it requires some human labor, and offers lower appeal in terms of innovation.
  • Autonomous robots bring a sense of innovation and futuristic appeal to retail environments and eliminate labor-intensive data collection. They can provide automated coverage across large stores, and studies show almost 80% of shoppers view robots in stores favorably. Key considerations for robot deployment include store coverage (narrow or blocked aisles), robust connectivity, ongoing reliability, and maintenance. For large stores 1-2 updates per day are the upper limit due to speed and battery size, and multiple robots are usually uneconomical.
  • Fixed cameras excel in reliability, require no labor, and offer consistent connectivity. Their power supply means charging isn't required, and they have a wired internet connection to store systems. Once deployed, they enable a retailer to take images at any frequency (daily or hourly) at no additional cost. However, initial deployment can be a significant investment and effort consideration, often requiring professional installation and calibration.
  • Wireless cameras, like fixed cameras, offer advantages in the areas of reliability, labor use, and capture frequency scale. They excel in enabling quick and easy implementation across many store formats or shelves. Key considerations include regular battery replacement and connectivity.

A hybrid data capture approach

While each retail data capture method offers unique advantages, a hybrid model, combining multiple methods within the same store or across different stores in a retail chain, is often the most cost-effective and flexible approach to shelf intelligence.

It allows retailers to tailor shelf intelligence to their specific operational needs.

For example, retailers can start with existing mobile devices for a quick pilot, minimal investments, and rapid results. Once performance is assessed and benefits proven, retailers can add fixed/wireless cameras in key aisles or robots in larger stores to enhance automation and efficiency, maximizing return on investment (ROI).

This approach allows for:

  • Comprehensive coverage of all store areas and different store types
  • Balancing of cost and efficiency
  • Flexibility to adapt to changing needs

Hybrid data capture shelf intelligence case studies

Here are two anonymized case studies that demonstrate how a hybrid data capture solution can be leveraged in various ways to meet the diverse needs of retailers.

Large European grocer

Goal: Improve on-shelf availability in medium-sized supermarkets and large hypermarkets. Generate high-value data for monetization with CPG vendors.

Data capture needs: Daily updates for critical categories, with less frequent full-store audits.

Solution:

  • Phased approach using mobile data capture for initial deployment, enabling flexible, targeted audits and rapid issue resolution.
  • Fixed cameras added to high-value, high-traffic hypermarket shelves for frequent KPI monitoring, while mobile devices stay in play for rapid scaling.
  • Approximately 80% of hypermarkets and supermarkets are well-suited for robots, representing the next step in scaling automation.

Result:

  • Achieved over 95% on-shelf availability and planogram compliance, resulting in 2% sales uplift within 6 months.
  • Balancing CAPEX and OPEX reduced total costs while enabling a strategic path to increase automation.
  • Category-level insights are now packaged for CPG data monetization.

Large North American drugstore chain

Goal: Improve on-shelf availability and increase labor efficiency. Address inventory inaccuracies and ensure visual proof of task execution and compliance.

Data capture needs: Weekly shelf capture due to low delivery frequency.

Solution:

  • Economics: Mobile capture is the most economical option, leveraging the retailer’s existing Zebra devices.
  • Low frequency: With a weekly update cycle, robotics and fixed camera systems are not cost-effective.
  • Store layout constraints: Narrow aisles make the use of robots impractical, and fixed/wireless cameras struggle with field of view limitations in smaller stores or specific areas.
  • Potential to add fixed cameras in larger stores on fast-moving shelves to increase automation.

Result:

  • Identified 43% workable out-of-stock to increase on-shelf-availability by +5.0%pt.
  • Existing TC52 hardware is proven to be compatible with shelf intelligence data capture for on-shelf-availability and price compliance use cases.
  • Full-store capture in under one hour using existing hardware, boosting employee satisfaction, and overall operational productivity.
  • Planned elimination of gap scanning and price check processes.
  • Elimination of significant phantom inventory i.e. items marked as in stock but unavailable. This led to immediate balance-on-hand corrections in the store system and quicker reorders, improving inventory accuracy - leading to increased on-shelf availability and sales, with no additional capital expenditure.

Key considerations when choosing a shelf intelligence solution

After selecting the optimum retail data capture methods, the next step is to choose the right vendor. While understanding each method's capabilities is important, finding the right shelf intelligence provider is essential. Here are six key considerations:

  1. Accuracy is critical: The shelf insights accuracy* provided to associates must be higher than 99.5%. Anything below that threshold undermines efficiency, and user acceptance, missing revenue-generating opportunities, and failing to prevent out-of-stocks.
  2. Data capture offering: The ideal vendor should offer compatibility with different data capture methods (e.g., mobile, fixed/wireless cameras, robots) that can be tailored to different store environments and data collection frequencies. It ensures high ROI, balances CAPEX and OPEX effectively, and makes it easier to scale across diverse store types without compromise.
  3. Vendor experience: Selecting a vendor that understands retail workflows, has robust data privacy and security standards, and domain-specific models can help navigate the process of change and achieve faster time to value.
  4. Systems integration: Shelf intelligence must work in harmony with existing technology stacks, encompassing planogram systems, inventory management, and task execution tools. Look for vendors that offer robust APIs, integration capabilities, and experience working with these systems.
  5. Actionability and prioritization of insights: It’s not just about collecting data; it’s about acting on it. Ensure the vendor can deliver insights in a clear, actionable, and tailored format with prioritization built in, so store teams can action the ones that matter most.
  6. Scalability and support: Vendors should offer a solution that can easily and quickly scale across formats and geographies, and provide ongoing support and innovation as needs evolve. Ask about post-deployment support, roadmap visibility, and a partnership approach.

* Shelf insight accuracy is calculated as the opposite of the % of incorrect alerts for all SKUs scanned.

Key takeaways

Choosing the best retail data capture method depends on a retailer’s operational goals, store format, and budget considerations.

Each method—mobile, robotic, fixed, or wireless cameras—has its strengths and weaknesses. Retail decision-makers should evaluate the accuracy, labor, and cost trade-offs for their specific stores.

In many cases, a hybrid approach is the most cost-effective and flexible way to deploy a shelf intelligence solution.

By carefully assessing their specific requirements and constraints, retailers can design a customized shelf intelligence system that maximizes on-shelf availability and drives business success while keeping investment at a minimum.

Illustration of machine learning models trained to recognize packaging or products on shelf.

Actionable shelf intelligence

Insights you can trust powered by hybrid data capture