Cd can be gotten easy enough if you have a level "runway" and a means of measuring time and speed. See the
Instructables "Measure the drag coefficient of your car" page. I have done the coast down testing like they do a couple times, and have modified the simple spreadsheet to meet my needs.
My findings can be
seen here.
50% grill block + windshield wiper block + folded passengers mirror + smooth underpanneling under the engine compartment + smooth underpanneling behind the rear axel = took my Cd from 0.32 to 0.28 (a 12.5% reduction)
All of the above (except the passengers mirror was folded back out) + 4' kammback (on a sedan) = took my Cd from 0.28 to 0.245 (another 12.5% reduction)
Things to be aware of when doing coast down testing, you need to do several bi-directional runs and average the results. I have been doing 5 runs in each direction, ASME recommends 10. Also, to get accurate results, you need both high speeds (to get the Aero) and low speeds (to get the rolling resistence), so if you can't do a 60mph to 15mph run in one pass, you may have to do 60-30 and 35-10 (overlap is ok), stitch them together and run the spreadsheet. If you need help figuring out how to make it work, I would be happy to give pointers.
I believe coast down testing is the method used by basjoos to determine his Cd, and I think aerohead has used it to gauge impacts also. Not everyone has run it through the spreadsheet as I have, but it is easy to overlay the graphs on top of each other and see that you are coasting longer/faster.