Lexical development is an indispensable component in child language studies. Familiarity with it's features makes it possible to recognize deviations in the context of various communication, language and speech disorders, such as developmental language disorder. It is determined by various factors, including phonotactic probability and neighborhood density. This paper contains an overview of some research about these phonological/phonotactic factors, their influence on speech and language processes and lexical development. It also includes a case study based on transcripts of a girl from the Croatian Corpus of Child Language. The aim was to obtain data on phonotactic probability, neighbourhood density and clusters of words she uttered early during development (1;3 – 2;8). Nouns and verbs were extracted from the transcripts using the CLAN program and they were analised in two phases of language development – premorphological and protomorphological. The Phonological CorpusTools program was used to calculate their phonotactic probability and neighborhood density. Furthermore, words (nouns and verbs) with consonant clusters were singled out and their preferability was determined using the NAD Phonotactic Calculator. Percentage of clusters in dense and sparse words was also calculated. Hypotheses assuming higher percentage of words with high phonotactic probability and neighborhood density were partially confirmed, and there is also more words with preferred clusters, compared to words with low phonotactic probability, neighborhood density and those with non-preferred clusters. In the premorphological phase clusters are more represented in high density words, while in the protomorphological phase they are more represented in low density words.