Adaptive Price Optimization: Leveraging Google Cloud AI for Real-Time Retail Transformation
Ganesh Vetsa, Sai Bhaskar Reddy Kovvuri, Pavan Kumar Kanamarlapudi, Balaji Barmavat, Krupa Satya Prakash Karey, Iwin Thanakumar Joseph S · 2025
Dynamic pricing optimization is a groundbreaking strategy that enables businesses to enhance sales, improve customer experience and stay competitive in rapidly changing business environments. Conventional pricing strategies that employ static models and historical data frequently does not have the ability to vary instantaneously according to the action of pricing strategies, or customer behaviour. This research article presents an end-to-end solution utilizing cloud-based artificial intelligence with Google Cloud technologies such as Vertex AI, BigQuery, and Cloud Run, to carry out real-time pricing optimization. Various ensemble-based machine learning prediction models, like Random Forest, XGBoost, LSTM and ARIMA time series models are used to predict prices and make real-time pricing predictions. This research article also highlights how the scalable, reliable, efficient and accurate system provide decision-makers with the capability to capture and incorporate observational data to establish reliable, data-driven pricing policies. Random Forest model performed best with the highest dynamic price prediction accuracy of 98.20% and mean absolute percentage error (MAPE) of 2.79 while compared with other state of art methods like LSTM, xgboost and lightGBM because it can learn intricate temporal relationships in price data.