A simple diphasic approach was applied to assess the self-compacting ability of concrete incorporating flash-metakaolin as the sole mineral admixture. The approach was based on tools available in the literature. At cement paste scale (cement + metakaolin + water + HWRA), interactions among all the constituents were considered by means of the wet packing method. Starting from the water volume just necessary to fill the voids between suspended powder materials at maximum packing density, an excess water volume was deduced to adjust the paste flow for self-compacting applications. At aggregate scale (sand + gravel), the water volume retained by the projected granular skeleton (maximum packing) was demonstrated not to alter the flow properties of the paste. The paste/aggregate association was tested with Ordinary Portland Cement (OPC), blended cements (limestone or slag), polycarboxylate-type high range water reducing agent (HRWRA) and rounded siliceous sand (0/4 mm) and gravel (4/14 mm). Results showed that the metakaolin did not act as a filler, and induced a decrease in the wet packing density because of its particle size, which was greater than or similar to that of the cement particles. In addition, the BET surface area was a relevant criterion for the optimization of the paste flow. With regard to industrial applications, viscosity measurements could be simplified by the Marsh cone flow test. Results also showed that the paste/aggregate association initially tested and validated with limestone fillers could be adapted to the design of self-compacting concrete incorporating metakaolin. The viscosity criterion determined for proper flow of paste incorporating limestone filler remained true provided that the mixing time was prolonged at the stage of the rheological characterization of cement/metakaolin suspensions.