The French Data Protection Authority (CNIL) has fined Amazon France Logistique €32m over its ‘excessively intrusive’ system for monitoring employee activity and performance.
According to Euronews, the fine is the equivalent of about 3% of the company’s annual revenue, which reached €1.1bn in 2021.
The watchdog also fined the business, which manages Amazon group’s large warehouses in France and employs 20,000 people, ‘for video surveillance without information or sufficient security’.
The French authority says it undertook several investigations into Amazon France Logistique’s practices in its warehouses after being alerted by articles in the press and receiving some complaints from employees.
As the CNIL’s press release explains, each warehouse employee is supplied with a scanner to document the performance of certain tasks that are assigned to them in real time, such as the removal of an item from the shelves or packing.
When they scan tasks, this data is recorded and then stored. Significantly, it is also used to ‘calculate [three] indicators that provide information on the quality, productivity and periods of inactivity of each employee’.
However, in its ruling on 27 December, the CNIL said that it was illegal for the business to set up a system that measured work interruptions with such accuracy, as this potentially required ‘employees to justify every break or interruption’.
The authority cited Amazon France Logistique’s ‘idle time’ indicator, which signals periods of scanner downtime of ten minutes or more, as well as its ‘latency under ten minutes’ indicator, which signals periods of scanner interruption between one and 10 minutes.
It also said that the system for measuring the speed at which items were scanned was ‘excessive’. This relates to the company’s ‘Stow Machine Gun’ indicator, which signals an error when an employee scans an item too quickly (i.e. in less than 1.25 seconds after scanning a previous item).
The authority also considered that it was ‘excessive’ for Amazon France Logistique to keep all the data collected, as well as the resulting statistical indicators, for all employees and temporary workers, for 31 days.
In its ruling, the CNIL said that the stock and order management process in the company’s warehouses broke down into several tasks, such as receiving items, storing inventory, preparing and sending orders. To carry out these tasks, each employee needed management to provide them, if necessary, with assistance or to reassign them to other tasks.
However, it added that providing this assistance or reassigning employees in real time did not require access to ‘every detail of the employee’s quality and productivity indicators collected using the scanners over the last month’.
THE INTRODUCTION OF WAREHOUSE MANAGEMENT SYSTEMS AND SIMILAR TECHNOLOGIES NEEDS TO BE RISK ASSESSED AND MEASURES INTRODUCED TO ENSURE THAT ANY EMPLOYEE WELFARE RISKS THAT MAY BE ASSOCIATED WITH THEIR USE ARE ACTIVELY MANAGED.
The CNIL pointed out that supervisors could already rely on data reported in real time to identify any difficulties an employee may be experiencing that may require assistance or to identify employees who could be reassigned to a task in the event of a peak in activity.
The authority concluded: ‘In addition to real-time data, a selection of aggregated data, on a weekly basis, for example, would be sufficient’.
In its press release, the CNIL also noted that the business had failed to properly inform employees and external visitors of the video surveillance systems used.
It also said that access to the video surveillance software was not sufficiently secure because the access password was not strong enough and the access account was shared between several users
‘This accumulation of security defects makes it more difficult to trace access to video images and to identify each person who has carried out actions on the software,’ the authority said.
As the CNIL explains, the size of the fine reflects the fact that the business’ decision to process employee data using scanners was different from traditional activity-monitoring methods due to the scale on which they were implemented. This, it added, led to very close and detailed monitoring of the work that employees did.
‘Such systems kept employees under close surveillance for all tasks carried out with scanners and thus put them under continuous pressure,’ said the CNIL.
However, Amazon has challenged the CNIL’s decision and said that its conclusions are factually incorrect. The business says it reserves the right to file an appeal.
‘Warehouse management systems are industry standard and are necessary for ensuring the safety, quality, and efficiency of operations and to track the storage of inventory and processing of packages on time and in line with customer expectations,’ Amazon said.
Its statement added that Amazon uses technology to support its employees, ensure the safety and quality of its operations, and to help the business deliver for its customers. The warehouse management systems like the ones discussed in the case are common in the industry, Amazon said, noting that it had worked closely with CNIL throughout the case to address its concerns.
Commenting on the Amazon case and its wider implications for businesses that may be considering workplace surveillance, Ian Manners, partner at legal firm Ashfords LLP, told IOSH magazine: ‘While it is inevitable and understandable that organisations will look to utilise technologies, including warehouse management systems, in order to identify areas of process optimisation, they need to be careful when doing so not to overreach into what may be argued to be excessive monitoring of the workforce.
‘Employers may argue that such systems are necessary for ensuring the safety, quality and efficiency of operations. However, they also need to recognise that their legal duty under the Health and Safety at Work Act requires them to ensure as far as reasonably practicable, the welfare as well as the health and safety of their employees. Consequently, the introduction of warehouse management systems and similar technologies needs to be risk assessed and measures introduced to ensure that any employee welfare risks that may be associated with their use are actively managed.’