Vortex generators are known to improve lift on aircraft wings, enabling lower takeoff and landing speeds, better climb rate, while reducing stall speed. So, using this technology, for a given acceptable stall speed, they would enable an otherwise smaller wing, which in turn would mean higher cruise speed. This happens due to better attached flow over the wing.
Julian Edger shows some improvement in fuel economy, attached flow on the rear glass and deck, and perhaps less noise. Noise is a good indicator of turbulent and detached flow, so less noise would be consistent with smoother air flow.
Vortex generators can be make for virtually no cost: Using two-sided adhesive tape such as carpet tape, and bent bits of aluminum, one could make small VGs out of, say, a used beer can. Just cut ~1" X ~1" rectangles, bend 90 degrees at centerline for an L shape, then trim one corner (which will be the leading edge) to rounded shape with a fingernail clipper. (If a beer can skin is too thin, use thicker material from a hardware store such as aluminum ducting for furnaces and chimney flues, etc.) Attach two-sided tape to the untrimmed side of of the L, and stick to the car surface at ~15 degrees off of flow direction, making V shapes with sets of VGs.
Google (
http://www.microaero.com/ImageGaller.../ig_index.html) for vortex generators for airplanes, and see how they are attached at normally turbulent flow areas, such as inside the angle of horizontal and vertical fin, wingroot, etc.
On a car, one might imagine tuft testing of VGs where ever there is detached flow, such as various places on the hood, roof, rear glass, in front of and behind wheel wells, on undertray (Securely attached GoPro camera would be good for this.), etc..
So, for less than a dollar's worth of carpet tape, a used beer can, some yarn tufts and bits of scotch tape to attach, and with a buddy in a chase car with a camera, you could do all sorts of testing.
Consider: Edger's 3% fuel economy improvement may not be much in itself, but then he only applied VGs to a tiny portion of his car--the rear window. What would happen if he'd gone whole hog and tested
everywhere the tufts showed bad flow? One might imagine 3% here, 2% there, 1% someplace else, and pretty soon we're talking about significant fuel savings.