In a distributed data processing environment, data skew can significantly impact performance. Describe a scenario where you encounter data skew in a Spark job and outline the steps you would take to mitigate this issue. Include specific techniques or configurations you would apply to balance the data distribution across nodes.