[DoctorDomo]

Lets use proton NMR as an example, and lets say all protons are -O-H protons so that only singlets are produced. If your spectrum has more unique protons than your product could possibly have, its obvious you have a mixture of products, but I was wondering if you can still use this. For example, with the software could you overlay the spectra of your precursors and then make any overlapping peaks invisible?

[Corribus]

You can DO anything you want. The question is whether you SHOULD.

In this case it would depend a lot on what you're trying to show.

[DoctorDomo]

Lets say you have traces of starting material left in your product which is impractical to remove. Can you just overlay the spectrum of the pure starting material with that of your product in order to make the starting material invisible? The only problem I see is that the starting material might have some identical chemical shifts to the product, so you'd end up with a partial spectrum, but having a partial spectrum would still be useful because it'll consist of peaks corresponding to new functional groups, and old functional groups which have had their electronic environment altered significantly. Another problem is that some of the new or shifted peaks might overlap with the old peaks from the starting material, so some new functional groups would be made invisible. Still, a partial spectrum could reveal useful information.