A practical guide to benchmarking entity resolution (ER) systems. It covers commonly used public datasets, explains which evaluation metrics are informative in ER (and why accuracy can mislead), and outlines how to design a domain-specific test set so results are meaningful for production decisions.