The aim of this work was to elaborate a novel approach for the development and optimization of self-emulsifying drug delivery system (SEDDS), using response surface methodology and hydrophilic lipophilic balance (HLB). Fenofibrate was selected as drug model. Rapid selection of excipients was operated with reference to their toxicity and capacity to solubilize the drug. A three level Box–Behnken design was used. The independent variables were (X1) surfactants/oil, (X2) cosurfactant/surfactant and (X3) percentage of cosolvent. The high and low levels of these factors were selected with reference to the experimental domain that covers an interval of HLB from 7.8 to 15. This interval of HLB is assumed to lead to oil in water emulsification and to contain the required HLB. The responses were (Y1) droplet size and (Y2) cumulative percentage drug released in 20 mn.Various response surface graphs were constructed to understand the effects of different factor level combinations. The optimized SEDDS with predicted drug release 83.6%, and droplet size 137 nm was prepared; the experimental values were in close agreement. The required HLB was 9.85. Optimized SEDDS showed significant increase in dissolution rate compared to conventional prepared gelatin capsules. In conclusion, this paper demonstrated the reliability, rapidity, and robustness of the approach.