Electrical load forecasting is an indispensable concept in the planning, operation, and maintenance of the power grid. Forecasting electrical load is crucial from both a technical and financial perspective as it enhances the efficiency, dependability, safety, and stability of the power system. Additionally, it contributes to the reduction of operational costs associated with electricity generation and distribution. The present study discusses the electrical load forecasting of a distribution company in Karnataka. Predictive analysis is carried out using long short-term memory (LSTM)-based recurrent neural network (RNN) to forecast electrical load. As the load input data has missing values, the sequence is disturbed. Implications and analysis under such incomplete past data are analyzed. Libraries of Python programming language are used for electrical load forecasting. The outcome of the electrical load forecast shows that the training of RNN model fails with missing load data. Under such circumstances, the input missing data if replaced with ‘zero’ values resulted in huge error indices. The mean absolute percentage error (MAPE) in the present case is 5.06. This error could be further reduced by proper data management while training.