How Evri is using AI to make parcel delivery more efficient and personal
Evri is rolling out the latest Microsoft AI tools to save time, improve quality, and give customers more control over how parcels arrive.
Imagine a parcel arriving where you actually are, not where you hoped you might be when you placed the order. It could meet you outside your morning coffee shop, or at the supermarket while you’re charging your electric car.
That is the kind of service Marcus Hunter, chief technology officer at Evri, believes AI could help create: not by replacing the practical work of moving parcels, but by making it more visible, predictable, and responsive.
“Everybody talks about personalisation,” Hunter says. “But we’ve never had the tools to do personalisation properly before. Now we do. So over the next few years, we will eventually be able to give individually tailored, personalised deliveries to our 25 million customers.”
Avoiding ‘unhappy paths’
Parcel delivery looks simple from the outside: a person orders something, a courier brings it to the door and the customer gets an update. In reality, Evri runs a network that flexes around peak demand, with more than 30,000 couriers delivering every day and 6,000 to 10,000 more recruited during peak periods, such as Christmas or Black Monday.
Evri Group in numbers
Evri was founded in Yorkshire in 1974 and merged with DHL eCommerce UK in 2025 to form the Evri Group.
It delivers 1 billion parcels and 1 billion business letters a year to 25 million households and businesses across the UK.
It has 12,000+ employees, 30,000+ self-employed couriers, 8,000 delivery vehicles, 12,000+ ParcelShops and Lockers, and 500+ hubs and depots.
More info: www.evri.com/news-and-corporate
When things go wrong – damaged labels, mis-routed parcels, or customers missing the update saying the parcel was delivered to a neighbour – these “unhappy paths”, as Hunter calls them, need addressing fast.
“I know it sounds easy delivering a parcel, but there are lots of complexities in this,” he says. “If something is damaged, what are we doing about it? How can we support a great outcome? And how can we help the customer if it’s an urgent package?”
Digital transformation
Evri has been building towards this for several years. In 2018, it moved more technology into cloud environments so it could flex capacity as demand changes. Around three-and-a-half years ago, Hunter says, it began testing AI and automation.
One early example was customer service. Instead of colleagues entering the same information into several systems, automation helps populate records, create notes and produce a summary. Hunter says that work, using Microsoft technologies to support automation in the Contact Centre, saved 500,000 hours and helped keep more service work in the UK.
Evri has also used Microsoft Power Apps to build a fleet tool tracking van size, routes, servicing and MOT dates. Hunter says it saved “hundreds of thousands of pounds” in avoided costs.
Across three years, he says Evri has saved around £34 million through automation and AI.
“We have a value case on everything that we do,” he says.
Controlled roll-out
The next step sees Evri implementing Microsoft 365 E7 – Microsoft’s latest product suite, which includes Microsoft 365 Copilot, Agent 365, Microsoft Entra and Microsoft Defender. Hunter wants to offer employees useful AI tools, but in a measured way that gives him a central view of AI agents, identity and governance controls, and security protection for agents that may be working when people aren’t.
As part of this pioneering move, Evri is rolling out 6,000 Microsoft 365 Copilot licences to its back-office workforce by the end of September, starting with finance, then procurement and legal. Copilot can help with everyday work such as drafting, summarising and finding information.
“What I didn’t want to do is just buy everybody a Copilot licence and say, ‘fill your boots’,” he says. “We want to do it in a methodical way.”
Digital team-mates
Hunter wants to avoid a workplace full of separate bots, many of which might be duplicative. Instead, Evri is planning five “super-agents”: courier, client, customer, corporate user, and operations.
Evri already has ‘Hey Charlie’ for human resources tasks in Teams, and ‘Eva’ for technology. Future uses could include checking contract clauses, helping employees book annual leave, supporting customer queries, and giving clients clearer information about delivery status, for example.
“I didn’t want to clutter Teams with 1,000 agents going, ‘Oh my God, which agent do I ask for what?’” Hunter explains. “If we have these five super-agents that also work together, it demystifies all the complexity.”
Better evidence
AI is also helping Evri understand what happens to a parcel in transit. At the distribution hub, parcels go through automated scanning, with several photos taken. At delivery, AI can help check whether the front door was open, whether the parcel was in the right place, and whether it matched the customer’s safe place preference.
For customers, that could mean clearer updates when a delivery changes route.
“It’s getting that information to the customer quickly and allowing time for them to read that as well,” Hunter says. “It’s really focusing on preventing those unhappy paths.”
Hi-tech future
The longer-term vision is a service less dependent on one standard process. If customers set a preferred safe place in the Evri app, with a photo and precise location, Evri can use that information and build on successful deliveries.
“Customers can support us with more information, by highlighting their safe place, for example” Hunter says. “AI will help us communicate more clearly.”
He also sees AI, automation and robotics changing the “middle mile”, the journey from a central hub to a depot. His example is Rugby to Carlisle, a trip that might take a lorry three-and-a-half to four hours, but could take around half-an-hour by unmanned drone.
That depends on regulation, infrastructure, and testing. But Hunter believes the roadmap is clear: more choice for customers, better visibility for colleagues, and more consistent services across rural and urban areas.
“You’ll see more efficiencies,” he says. “And with more efficiencies, you’ll see more personalisation and a better service for customers.”