We need to understand (explore) the data at the source of topological uncertainty/ ambiguity, the why of the "weak spots" in trees. In particular, when the aim should be cladistic (clade-based) or a relatively stable phylogenetic (sensu Hennig) classification.

Inferring trees and applying methods eliminating (post-analysis weighting) or masking (Bayesian inference) internal signal conflict, may not be the best choice, when one has to rely exclusively on (highly) homoplastic morphological data that provide little tree-like signal. In contrast, networks can visualise incompatible signals and competing topologies or varying support in MPT and bootstrap pseudoreplicate (or Bayesian-inferred) tree samples. Thus, tell us directly where our data are unequivocal, and where it support several equally valid alternative scenarios.