Group relations between languages, especially cognate words, provide an excellent opportunity to develop receptive competence (“intercomprehension”). This paper presents a computerized approach to the investigation of the extent of Dutch–German cognates in Dutch and the difficulties a German reader might have in recognizing them. The main procedure used to finding the most similar German counterparts of Dutch words is based on the Levenshtein algorithm, in which findings and assumptions on general similarity perception have been integrated along with a set of statistically important sound correspondences. Results show that about 75% of the tested frequency list of 5,000 Dutch word forms can be decoded with the help of German.