Standardised tests of whole-word accuracy are popular in the speech pathology and developmental psychology literature as measures of children's speech performance. However, they may not be sensitive enough to measure changes in speech output in children with severe and persisting speech difficulties (SPSD). To identify the best ways of doing this, we compared a range of commonly used procedures for perceptual phonological and phonetic analysis of developmental speech difficulties. Data are drawn from four children with SPSD, recorded at 4 years and again at 6 years old performing naming and repetition tasks. Measures of speech output included percentage of whole words correct (PWC), percentage of consonants correct (PCC), proportion of whole-word proximity (PWP), phonological pattern (process) analysis and phonetic inventory analysis. Results indicate that PWC captures change only when identical stimuli are used across time points. PCC is a more robust indicator of change, being less affected by the choice of stimuli. PWP also captures change across time and tasks, while appearing to be more sensitive than PCC to psycholinguistic variables. PCC and PWP are thus both potentially useful tools for evaluating speech outcomes.