Monarch butterfly wings are covered in minuscule scales (approximately 100 µm in length) which align together in a pattern that resembles roof shingles. Flight tests performed on live butterflies showed that these scales provide a beneficial aerodynamic function. The scales are angled upwards such that transverse cavities form for a flow passing perpendicular to the rows of scales. Flow visualization of butterfly inspired cavities has shown that the entrapped vortex (or vortices) inside each cavity can act as a fluidic bearing to the outer boundary layer flow resulting in reduced surface or skin friction drag. This study conducted experiments on cavity embedded flat plates, mimicking the butterfly scale geometry, documenting this “roller-bearing effect”. The experiments were performed in a tow-tank facility, and DPIV measurements were used to calculate the surface drag based on the measured boundary layer velocity profiles. Initially, a net skin friction or surface drag reduction for a simple rectangular cavity of aspect ratio (AR) 2 was measured. However, as the cavity geometry was varied to better mimic the butterfly scales, the shape of the vortex and the dividing streamline also changed, which in turn affected the surface drag. This study aimed to determine the cavity geometry that results in the highest drag reduction. Experimental models with cavity geometries of AR 2 and 3 and wall inclination angles of 22°, 45° and 90° were tested, where the slanted models were inspired from observed butterfly scale geometries. Each of these models exhibited net surface drag reduction for lower Red range of 4.5 to 8.5, except for the 22° cavity with AR 2 which showed an increase in drag for all the Red cases. The Red is defined based on the cavity depth which is kept constant for all the models. The model with a 45° cavity wall inclination and AR 2 had the highest surface drag reduction ranging from 18.63% to 26.33% at lower Red range. Additionally, the critical Red region, beyond which the embedded vortex becomes unstable, and fluid begins to enter and eject out of the cavity, mixing with the outer boundary layer flow and thereby eliminating the drag reduction effect, is documented for various cavity geometries and ranged from Red 90 to 120. As was hypothesized, the increase in Red beyond the critical region resulted in an increase in surface drag.