There’s a bottleneck hiding inside most electrified last-mile fleets, and it has nothing to do with range, charger availability, or electricity costs. It’s the gap between the charging system and the dispatch system. One knows how much energy each vehicle has. The other knows when each vehicle needs to leave and where it’s going. In most fleet deployments, these two systems don’t talk to each other.

Charging infrastructure for last-mile delivery fleets tends to get owned by facilities or energy teams. That makes sense during procurement and installation. But once vans are running daily routes, charging stops being an energy question and becomes a logistics one. Your route optimization engine decides which vehicles leave when, with what payload, on what timeline. Your dispatch system handles multi-shift handoffs where a van finishes its morning route and needs to be back out by early afternoon with a different driver and a different set of stops. These systems define when vehicles are available and when they’re needed. But the charging system, in most deployments, doesn’t have access to any of that information.

So what actually happens is this: someone on the operations team cross-references the dispatch schedule with the charging dashboard, manually adjusts charge sessions, and makes judgment calls about which vehicles to prioritize. At ten vans, that’s a minor inconvenience. At a thousand, across twenty depots in different states with different utility rate structures, it’s a structural bottleneck baked into every shift change.

What This Costs at Scale

The costs aren’t dramatic. There’s no single catastrophic failure. Instead, it’s a steady accumulation of small inefficiencies that compound as you grow.

Vehicles depart late because they weren’t charged in priority order. The van with a 2 PM pharmacy commitment got the same power allocation as the one not dispatched until tomorrow morning. Infrastructure sits partially idle during off-peak windows because load management can’t dynamically shift capacity based on the next departure schedule. Revenue from grid flexibility programs goes uncaptured because participation requires automated, bidirectional coordination between your charging system and an aggregator, and that integration was never built. Meanwhile, the engineering team maintains a fragile middleware layer between TMS and CPMS that breaks with every vendor software update, consuming development cycles that should go toward the core logistics product.

None of these problems show up in the pilot phase. They emerge at scale, and by then they’re embedded in the operational assumptions of the whole fleet.

Treating Charging as Part of the Logistics Stack

The fix is architectural: the charging system needs to sit inside the logistics stack, not alongside it.

Concretely, that means the charge management platform pulls directly from your WMS and TMS through API integration. When a route plan changes or a vehicle returns early, charge scheduling adjusts automatically without anyone opening a second screen. Power allocation follows delivery priority, not connection order. The van with the next departure and the most urgent route gets charged first, every time, without a human making that call.

For fleets operating across multiple states, this also means the system accounts for regional utility rate structures and demand charge thresholds at each depot, optimizing collectively rather than site by site. And for fleets thinking about V2G, a logistics-integrated charging architecture turns your parked delivery vans into distributed energy assets that can generate revenue during off-peak hours, which starts to meaningfully offset charging costs once the fleet reaches a certain size.

Why Most Platforms Weren’t Built for This

The majority of charge point management platforms on the market were originally designed for public charging networks, where the core problem is availability and payment processing for individual drivers. Fleet management features were added later, and they tend to address fleet-specific concerns (reporting, access control, basic scheduling) without deep integration into logistics workflows. That’s not a criticism of those platforms. They’re good at what they were built for. But last-mile delivery operations have a fundamentally different set of requirements, and stretching a public-network CPMS into that role creates the kind of friction that’s invisible at small scale and painful at large scale.

This is why a growing number of fleet operators are building a custom orchestration layer on top of the OCPP protocol rather than relying on an off-the-shelf platform to cover the full stack. OCPP handles charger-level communication reliably. That part is commodity infrastructure. The orchestration logic on top, the part that connects charging decisions to delivery priorities, is where the operational value lives. And it’s worth owning.

Where to Start

If you’re running a small fleet at a single depot, none of this is urgent. A standard CPMS configured well will get the job done, and speed of deployment matters more than architectural flexibility at that stage.

But if you’re planning to scale electric last-mile delivery across multiple regions, or you’re already feeling the friction of manual coordination between dispatch and charging, it’s worth asking a different question than “which CPMS should we buy?” The better question is: how does charging become a native part of our logistics operation, rather than a parallel system we manage around?