Currently, agricultural activity is a fundamental part of a country’s economy and food security, where rural areas play a crucial role in employment generation and local economic growth. For this reason, the ability to forecast agricultural production and sales is of utmost relevance for planning and decision making. This article reviews several studies related to agricultural forecasting and the techniques employed, such as machine learning algorithms and statistical models. The main objective is to provide a useful and beneficial tool for small and medium farmers, as well as for those responsible for agricultural areas. First, we proceeded with massive data collection that guarantees reliable results. Then, using machine learning techniques, we performed the analysis of annual sales and production for the last three years for a wide range of agricultural crops, including cereals, pulses, tubers, vegetables, and other agricultural products. This methodology facilitated the prediction of which crop will be the best in both production and sales and to which province and region it will belong next year. This will provide predictions and recommendations for growing crops in each annual season, which will contribute to making informed decisions, especially for retail farmers.