After a long and arduous attempt at an elucidation, it is quite common to be left with more than one candidate structure. In some cases collecting more data is not an option and an exhaustive database/literature search turns up nothing useful, the alternative approach is to synthesize the proposed candidates and then compare the spectral data to the original unknown.
The example below shows three proposed candidate structure differing in the attachment of the hexopyranoside group. The high degree of similarity between the candidates and the high number of quaternary carbons (11 per structure) makes it very difficult to narrow the list to one candidate.
When data is limited and thus cannot assist in eliminating two out of the three candidates, then the only option left is to synthesize the candidates and compare the spectral data.