Unfortunately, the chromatogram is very indistinct on my screen, more like looking through opaque glass, so I can't comment on the separation.
However, if the replicates are very variable, you have to sort out that problem first. Assuming that your retention times are constant, the areas of the small peak are presumably varying. If the large peak is also varying then you have even more problems to address.
The most obvious issue would be the integration, and you should have two sharp, well-defined peaks for the software to work with. 
If not, then you probably should first decrease the sample injection volume, but still ensuring your smaller peak stays above the lower limit of quantitation. 
Try to ensure your sample is dissolved in the mobile phase, and that you have nice peak shapes - eg play with column temperature etc to achieve the sharpest peak, perhaps try using a gradient, provided you have that option on your HPLC. Target about  8 - 12 mins retention of large peak for your column.
By increasing the organic component you should increase the retention, and may improve the separation ( depending on peak shape ) but you are already at 66% organic, so you have to be careful about issues such as solubility as well. 
You may have to change to a different column chemistry, preferably one that moves the small peak  either much further away, or in front of the large peak. 
Good luck,
Bruce Hamilton