Determine whether the small size of the sample truly influences your results in a negative way. Choose an effect size measure, and compute it. Some choices for effect size measures are Cohen's "d," Eta Squared and Omega Squared. For example, if your experiment is comparing two means, compute Cohen's "d" by subtracting the means and dividing by the pooled standard deviation. If this value is under 0.3, then your small sample size will likely pose a problem in this study. If the value is above 0.3, then you do not need to worry about adjusting for the small sample size.
Increase the budget of your study. If you found your effect size was too small, use this effect size to argue for a larger budget. If this study is important to your sponsors, the small effect size will likely warrant a larger budget, which will allow you to gather a larger sample.
Reduce the variance of your statistic. This can be accomplished through increasing the accuracy of your data-measurement. For example, if you are measuring physical objects, switching from human measurement to computer measurement can help reduce the error, and hence variance, related to the statistic. A high variance negatively contributes to the effect size, so finding a method to reduce the variance may alleviate the problem of a small sample, regardless of whether you increased your budget.