Logistics firms lag on AI despite delivery ambitions

Zebra Technologies research, conducted with Oxford Economics, found a gap between logistics companies' AI ambitions and their progress in improving delivery operations.

Among logistics leaders surveyed, 48% want to use AI to improve operational efficiency, while 34% are focused on customer experience. Yet 62% said delivery and field operations still need critical improvement and that they have not made meaningful progress in those areas.

The findings point to a tension in logistics as operators face rising pressure from eCommerce demand, more complex supply chains and tougher expectations around delivery performance. They also highlight a divide between boardroom plans for automation and the day-to-day reality for warehouse and delivery staff.

Phil Sambrook, Transport and Logistics Strategy Director, EMEA, said companies need to focus more closely on frontline workers if they want AI projects to deliver operational results.

"Bridging this gap is an opportunity for logistics leaders to elevate the role of the frontline, where customer transactions and trust are often won or lost," said Phil Sambrook, Transport and Logistics Strategy Director, EMEA, Zebra Technologies.

Operational pressure

Zebra linked the issue to delivery losses, delays and inventory problems, citing examples from misplaced parcels to theft. Sambrook said better data collection at the points where goods move through the supply chain is central to any broader use of AI in logistics.

"The chocolate manufacturer could trace every single stolen bar because of unique codes on each package, effectively turning every retail scanner into a security checkpoint," Sambrook said.

"Advanced data capture is an essential foundation for frontline AI," he added.

Zebra described a model in which data from scanners, cameras and sensors is processed on devices used by workers in warehouses, depots and delivery fleets. Often described as edge or on-device AI, the approach is intended to help staff identify problems in real time rather than after delays or losses have already occurred.

In practical terms, machine vision systems above conveyor belts can distinguish genuine blockages from false alarms, reducing unnecessary stoppages and keeping parcels moving through distribution centres, according to Zebra.

Other systems can check parcel condition as items move through the network, looking for damage, leaks or missing labels before they continue on their route or enter returns processing. Workers can also use mobile computers with AI tools to scan multiple barcodes in one action, while visual overlays on screens guide order picking and identify the contents of sealed boxes.

Culture and learning

Sambrook said technology alone would not close the gap identified in the research. Companies also need to invest in training, workplace culture and visible backing from senior management if they want staff to adopt new systems and use them effectively.

"Implementing AI is more than an IT upgrade. It requires changes to workplace culture and learning resources to ensure adoption and success," he said.

That emphasis reflects a broader issue across the logistics sector, where many digital projects depend on workers using handheld devices, scanners and software consistently under time pressure. Analysts have long noted that new systems can struggle if companies fail to explain their purpose clearly or provide enough training for staff expected to use them on the warehouse floor or at the customer doorstep.

Zebra said some logistics operators are already using AI-assisted inspection and proof-of-delivery tools. In delivery operations, picture-based proof of delivery can capture and validate drop-off images, read barcodes and remove personal data in a single process.

According to Zebra, those systems have cut proof-of-delivery workflow time by 55% per stop in some customer settings and reduced annual claims costs by 10% to 30%.

Zebra has also expanded its range of AI products this year, including software tools and AI agents aimed at frontline operations. Uptake is growing among logistics customers that are using or testing those systems, it said.

Zebra was recently ranked 10th among S&P 500 companies for AI readiness in a Wall Street Journal report, reinforcing its effort to position itself as a supplier of AI tools for frontline industries. But the central message from its latest logistics research was less about software launches than about execution inside transport networks.

"By delegating resource-heavy tasks to on-device AI, warehouse and delivery jobs can be made less of a grind and more of a craft," Sambrook said. "This elevates productivity and helps employees feel valued by their leaders and engaged in their work for customers."