*I took the liberty to extrapolate your Cartesian grid in a velocity regression,down to 88.5 km/h (55-mph) and looked at 112.6 km/h.
*General Motors Laboratory published a metric for a Delta-Cd/ Delta-mpg relationship decades ago,presuming a constant brake specific fuel consumption for the engine.
*At 88.5 km/h,a 10% drag reduction would equate to a 5% increase in fuel economy.
*At 112.6 km/h,the same 10% drag reduction would equate to a 6% increase.
--------------------------------------------------------------------------
*Taking your liters/100 km data,and using the General Motors metric,at the two velocities,the drag coefficient of the A-Class M-B would have to fall from Cd 0.30,to 0.255-0.235 in order to account for the fuel economy improvement registered by the flow measurement equipment.
*This constitutes a drag reduction of 14.8%-21.6%.
-------------------------------------------------------------------------
*From some of the empirical wind tunnel studies published by Wolf Heinrich Hucho,and others,this range of drag reduction would fall outside the standard deviation for features optimization to a vehicle forebody,including airdam and grille-block with contemporary automobiles.
-------------------------------------------------------------------------
*Personally,I would want to take a harder look at the fuel measurement technology.Consumer scrutiny may require a closer fit with respect to the drag/mpg relationship.
--------------------------------------------------------------------------
*The bumper is quite attractive and merits all the blood,sweat,and tears necessary to may a go of it in the market.