This paper aims to provide a guideline for identifying the most suitable online voting system under the given requirements and acceptable tradeoffs. We have selected twelve (more or less) well‐known online voting systems that rely on distinct cryptographic mechanisms for achieving security. For each of the systems, we summarized the key architectural and cryptographic ideas behind their design. Then, we analyzed the required trust assumptions for achieving the three most important security properties (i.e., verifiability, divided between cast‐as‐intended, recorded‐as‐cast, and tallied‐as‐recorded verifiabilities, privacy, and receipt‐freeness). To make a fair comparison, we did our analysis based on identical security definitions. Note that we selected wildly known and well‐accepted definitions, which are scheme‐neutral, to avoid any biases. Also, we discussed some of the most critical practical aspects of those systems, such as–the necessity for secure or anonymous channels, reliance on secure printer facilities and so forth. To facilitate the comparison, we suggested a unified naming convention for system elements based on their roles and functions. Then, based on the unified naming convention, we compared all twelve online voting systems for both the security properties and practical aspects. Finally, we summarized our observations regarding patterns and dependencies we observed, provided guidelines for selecting the online voting system, and gave recommendations regarding system design. We hope our work contributes to the online literature and facilitates the process of selecting the most suitable e‐voting system depending on the requirements of a specific election.