The dealer network was a shock absorber. Direct sales removes it.
What manufacturers inherit when the independent retail layer no longer carries inventory, local variance, service capacity and delivery exceptions.
A luxury EV manufacturer wanted the margin, customer ownership and brand control of a direct-sales model. The attraction was clear. The infrastructure normally sitting behind the dealer was not.
There was no independent network to carry stock variation, reconcile local pricing, absorb delivery exceptions or provide a pre-existing aftersales and parts system. Once the dealer stopped being the operating buffer, those interfaces moved upstream to the manufacturer.
at.Pointe was engaged to design the retail, pricing, supply-chain, aftersales and parts model together with the business case needed for board approval and implementation.
A conventional dealer does more than sell the vehicle. The network finances inventory, flexes local capacity, manages delivery variance, carries workshop resources, handles pricing exceptions and resolves a large number of small problems before they become an OEM management issue.
Direct sales can remove the independent margin layer, but it also removes that shock absorber. The manufacturer inherits the work even when the organisational chart has not yet assigned it.
The margin does not move upstream on its own. The operating variance moves with it.
The launch could not be built from separate retail, aftersales, logistics and finance plans. The economics interacted too directly.
at.Pointe built the market launch around a modular business case connecting the main operating assumptions, investment requirements and performance measures across the model. Pricing was tested against local market conditions. Retail footprint and volume assumptions drove inventory and delivery requirements. Aftersales and parts capacity were linked to the installed parc rather than treated as a later service project.
The purpose was not to produce a larger spreadsheet. It was to make the trade-offs visible in one place, so that management could see what happened to capital, break-even and service readiness when one assumption moved.
Size the physical network and local capacity around expected demand rather than copying the footprint of an established dealer model.
Build local price logic from market reality while keeping central control of the commercial model.
Assign stock ownership, logistics, handover and exception management once the independent dealer no longer carries them.
Model service capability, parts demand and technical support early enough that the installed parc could be supported from launch.
Connect investment, working capital, profitability, decision rights and management reporting into the same launch logic.
The integrated strategy and business cases received board approval. That created the basis to implement the model, but approval did not make the operating system real.
The work then had to move into execution: footprint decisions, supplier and logistics setup, pricing routines, systems, parts, service capability, local organisation and the governance needed to run them together.
One direct-sales concept did not produce one set of economics. Across the pilot markets, capital requirements, break-even paths and operating burdens differed materially.
Direct-sales economics should not begin with the dealer margin as if that margin were a cost that simply disappears. The dealer was being paid to carry capital, local capacity and variance as well as to sell the car.
Once that layer is removed, retail, pricing, supply chain, aftersales, parts and finance become one launch decision. Weak integration is no longer hidden inside the network.
That can be a better model. It is only cheaper or more controllable when the manufacturer has deliberately built the operating system that replaces the buffer.
Written from operating experience, not market commentary.