Background and purpose Respiratory insufficiency is a common symptom during the course of amyotrophic lateral sclerosis (ALS). The diagnostic workup may be challenging and includes a wide array of diagnostic measures. In this study, the aim was to analyze the relationship between hypercapnia‐associated symptoms, blood gas parameters and pulmonary function tests. Methods In total, 109 patients (56 women, 53 men, 62.4 ± 11.9 years) with definite, possible or probable ALS according to El Escorial criteria were included. All patients received either arterial blood gas analysis, nocturnal capnometry or both. Pulmonary function was assessed by spirometry and peak cough flow. Clinical symptoms potentially indicating hypercapnia were assessed using 17 dichotomous (yes/no) items. Results Of 109 ALS patients, 40 had hypercapnia. The highest accuracy and specificity for predicting hypercapnia was observed for dyspnea at rest (Youden's index 17%, 95% confidence interval CI 2%–34%; sensitivity 23%, 95% CI 9%–38%; specificity 95%, 95% CI 88%–100%). Daytime fatigue yielded the highest sensitivity of 58% (95% CI 40%–76%). Logistic regression for all assessed symptoms combined yielded an area under the receiver operating charteristic curve of 0.8 (95% CI 0.7–0.9). Compared to the clinical symptoms, forced vital capacity and peak cough flow showed higher sensitivity (70% and 87%, respectively) but lacked specificity (33% and 20%). Conclusion Evaluation of the presence of hypercapnic symptoms can be utilized to predict incipient respiratory insufficiency and should complement pulmonary function tests. Further studies are needed to validate specific questionnaires in this regard. No single hypercapnia‐associated symptom or pulmonary function test on its own seems sufficient to safely predict hypercapnia. In this prospective study in ALS patients (n = 109), the diagnostic value of clinical symptoms as well as pulmonary function tests (spirometry and peak cough flow) was investigated to predict hypercapnia measured by nocturnal hypercapnia and morning blood gas analysis. It was found that the evaluation of multiple clinical symptoms combined showed a good predictive power to detect hypercapnic patients. No single clinical symptom alone had sufficient sensitivity and specificity. Compared to the clinical symptoms, spirometry and peak cough flow showed higher sensitivity but lacked specificity. It was found that, especially in bulbar patients, spirometry measurements were not reliable. Therefore, based on our data, using a combination of clinical signs and objective respiratory function tests is recommended for early and reliable detection of respiratory insufficiency.