It is customary to use various data source as well as triangulation techniques to improve the validity of a research. My question is which one is more important?
I assume that what you mean by "triangulation" is the optimization of experimental design and/or the cleaning up of the raw data. If that is indeed the case, you need to do the "triangulation" first, before you can apply the same to another data set. In this sense, your original question is miscast. One is not more important than the other; it is the order by which you do them that is important. You optimize your experiment using one data set, alongside true positive and true negative data sets. Then If you are successful, you can find out how general your experimental methodology is by using another data set. HTH.