This study proposes an improved VADER (IVADER) lexicon-based classification model for multi-domain customer sentiment analysis. The model constructs a domain-specific dictionary based on the VADER lexicon and classifies customer reviews using this dictionary. The IVADER model involves data preprocessing, vectorization, feature selection using WordnetLemmatizer, and an enhanced VADER lexicon classifier. The model achieves high accuracy (98.64%), precision (97%), recall (94%), and F1-measure (92%) with reduced training time (44 seconds). The model is validated on multiple datasets, including a multi-domain sentiment dataset and the Sentence Polarity dataset, demonstrating its effectiveness in classifying customer sentiment across different domains. The IVADER model addresses domain heterogeneity by constructing domain-specific sentiment dictionaries, allowing precise sentiment classification by considering the different sentiment directions of words. The model's performance is evaluated using various metrics, including accuracy, precision, recall, F1-measure, specificity, and AUC. The results show that the IVADER model outperforms existing studies in terms of accuracy, precision, recall, and F1-measure. The model is also tested with smaller datasets, demonstrating its stability and effectiveness in multi-domain sentiment analysis. The study highlights the importance of domain-specific sentiment analysis in the digital marketplace and the potential of the IVADER model in evaluating customer sentiment and introducing new products in competitive online markets.This study proposes an improved VADER (IVADER) lexicon-based classification model for multi-domain customer sentiment analysis. The model constructs a domain-specific dictionary based on the VADER lexicon and classifies customer reviews using this dictionary. The IVADER model involves data preprocessing, vectorization, feature selection using WordnetLemmatizer, and an enhanced VADER lexicon classifier. The model achieves high accuracy (98.64%), precision (97%), recall (94%), and F1-measure (92%) with reduced training time (44 seconds). The model is validated on multiple datasets, including a multi-domain sentiment dataset and the Sentence Polarity dataset, demonstrating its effectiveness in classifying customer sentiment across different domains. The IVADER model addresses domain heterogeneity by constructing domain-specific sentiment dictionaries, allowing precise sentiment classification by considering the different sentiment directions of words. The model's performance is evaluated using various metrics, including accuracy, precision, recall, F1-measure, specificity, and AUC. The results show that the IVADER model outperforms existing studies in terms of accuracy, precision, recall, and F1-measure. The model is also tested with smaller datasets, demonstrating its stability and effectiveness in multi-domain sentiment analysis. The study highlights the importance of domain-specific sentiment analysis in the digital marketplace and the potential of the IVADER model in evaluating customer sentiment and introducing new products in competitive online markets.