UK – National Sports Retailer

In June 2018 ShopperTrak, part of Sensormatic Solutions, engaged with a national sports retailer in the United Kingdom to run a proof of concept across fifteen of their stores in the North West.

The retailer already utilised a beam counting solution, but this was supplied by multiple providers, meaning that there was no consistent solution across the estate. Store Managers had little faith in the data, and there was a lack of confidence in the results being provided by the system.

The aim of the proof of concept was to highlight the accuracy of the ShopperTrak system in comparison to the beam solution, and demonstrate the insights that can be obtained by using accurate footfall data.

Strategy & Solution Provided

ShopperTrak installed its system alongside the stores’ existing beam counters. This allowed for comparison of the data at the same time. The ShopperTrak data was then audited on an ongoing basis to ensure data accuracy.

As part of the proof of concept, it was agreed that the ShopperTrak Retail Consultancy team would complete a piece of analysis to compare the traffic data from the new and existing systems, and look at conversion trends in the stores. For this, the retailer provided ShopperTrak with sales and transactional data for all stores taking part in the proof of concept, with four of the stores also providing staff rota details for the Black Friday and Christmas trading period. This allowed the consultancy team to complete a much more in-depth analysis which focused on labour optimisation alongside shopper behaviour. 

Implementation

ShopperTrak installed its cameras in 15 stores. The proof of concept lasted six months, allowing for analysis of the stores’ performance over the busy Black Friday and Christmas   periods, and providing the retailer with the analysis in the New Year.

The challenge: Differing figures from the two systems

Following the installation of the ShopperTrak counting solution, the difference between the two counting technologies became immediately apparent. It was discovered that 80% of the existing beam systems both over and under-counted visitor numbers when compared to the newly-validated and accurately-counting (+95% accuracy) ShopperTrak system.  No single store’s beam traffic counted in line with the ShopperTrak system.

Furthermore, some stores deemed to be the busiest with the beam system were found to be the quietest with the ShopperTrak system, meaning that Sales Per Shopper (SPS) and conversion trends also ranked quite differently. With no single beam solution consistently over or under counting at the same rate, and the overall weekly traffic variance ranging from -36% to +28% against the ShopperTrak system, this made it impossible for the business to know how its stores were performing.

The over and under-counting was not only seen on a weekly level, but also on a daily basis and even during the busiest hours of the week. Store Managers realised they weren’t able to align their scheduled labour to real demand.

The inaccurate traffic data was not the only issue discovered with the beam solution. 40% of stores who took part in the proof of concept saw the beam systems go completely offline for several days, or saw data come through intermittently on some occasions. As well as having no visibility of visitor numbers, this also meant the stores would have no visibility of sales per shopper (SPS) or conversion performance.

Altogether, this demonstrated a clear problem with the beam counting systems.

Impact on KPIs and labour planning

As part of the proof of concept, the retailer provided ShopperTrak with labour data for four stores. Using this, the ShopperTrak Retail Consultancy Team was able to demonstrate that using inaccurate and inconsistent data for labour planning has a significant negative impact.

In one example, the Shopper-To-Associate-Ratio (STAR) using the ShopperTrak data was significantly different to that of the beam system. Due to the beam system undercounting in this instance, it indicated that there was sufficient staffing in place for the number of people in the store, but the ShopperTrak data demonstrated that this was not the case. This was further highlighted by the ShopperTrak system showing that SPS and conversion were at one of the lowest levels of the day. These metrics can be significantly impacted by using inaccurate traffic data for staff planning. Our analysis also showed a direct inverse relationship between STAR and SPS, conversion and SPS, and Average Transaction Value (ATV and STAR. When STAR is lower, SPS increases, as does conversion and ATV. This highlighted the importance of using accurate and reliable data.

Outcomes and benefits

With these large traffic variances occurring during peak trading times, the retailer had no way to accurately assess how these stores performed in comparison to one another, but it also meant that accurate staff planning could not be achieved as there was so little consistency in data trends.

In contrast, ShopperTrak actively monitors and manages its devices, so should they go offline for any reason, the issue would automatically be investigated by the Client Support Team. Any missing data would have been projected based on previous trends seen at site, and as the devices store the actual traffic counts, this would overwrite the projected data once the device is brought back online. Therefore, the stores would still have an indicator of their performance while the issue was investigated.

Thanks to this constant auditing, the ShopperTrak system counts with +95% accuracy.

Following this, ShopperTrak became the retailer’s preferred footfall provider, with the aim of using the traffic data in its new workforce management system to plan staff rotas – something that would have been unachievable with the beam data.

case study button