Background
Northwind is a mid-sized aerospace prime with a 60-engineer CAE team. They had run the incumbent commercial CAE stack — structural FEA, CFD and full-wave electromagnetics from one vendor, an explicit-dynamics crash code from another, and a small external-aerodynamics tool from a third — for over a decade. The annual licence cost ran into seven figures, with a $1.8M renewal up at the end of the quarter.
The challenge
The team's CAE methodology was mature and the licence stack was deeply embedded in their workflows. Replacing it would require head-to-head proof on production cases — not on demo workspaces. The customer's CFO was prepared to renew on momentum. Our job was to show the technical team something better and let the financial argument follow.
What we did
- Ran a paid pilot across twelve real production cases — five external aerodynamics, three crash and impact, two EMC, two coupled fluid-structure.
- Seven cases had measured wind-tunnel or wind-on-rig data; five had production telemetry. Acceptance criteria came from the customer's existing CAE methodology.
- Mesh budgets and time-step budgets were matched between Aether and the commercial stack. Solver settings were tuned independently for each tool.
- A forward-deployed engineer paired with Northwind's CAE lead for the duration of the pilot.
“We're not buying a tool that competes with the incumbent. We're buying a tool that makes the question of renewing irrelevant.”
The outcome
- Aether matched or beat the commercial stack on accuracy in ten of twelve cases, was within 3% on the other two.
- Campaign completed in 38% of the wall-clock time at iso-accuracy.
- Northwind retired the $1.8M renewal at quarter-end and consolidated to a small residual seat count for niche tools Aether doesn't yet cover.
- Budget freed was redirected to a new hypersonics R&D programme — now also running on Aether.
One detail worth calling out
The clearest wins were on the coupled fluid-structure cases. Aether solved fluid and structural together on a single mesh; the commercial stack required two solvers, an inter-tool import, and a manual time-step matching pass. The customer's existing methodology had compensated for this with a small offset that they were carrying as a known quirk; Aether made the quirk disappear and the rig agreement improved.