This paper presents a new dialectometric approach applied to the Breton language in the heart of Breton-speaking Brittany. It is based on data from the Nouvel Atlas Linguistique de la Basse-Bretagne Le Dû (2001). We process qualitative data using the Levenshtein algorithm which allows us to accurately measure and take into account the discrepancies or similarities between different pronunciations of a given word. This contribution aims to determine whether linguistic distance is caused by a frequent repetition of the same phenomenon or whether it is the outcome of multiple changes. Our first results suggest new ways of analysing Breton data.