Collect and execute similar operations together rather than performing them individually to reduce overhead and improve performance. This applies to entity creation, API calls, and state updates. Key optimization strategies: 1. **Batch entity additions**: Collect all entities during initialization and add them with a single `async_add_entities()` call instead of multiple calls 2. **Combine executor jobs**: Group related synchronous operations into a single executor job to minimize expensive context switching between the event loop and thread pool 3. **Use selective updates**: Choose appropriate update mechanisms - use `async_update_listeners()` when state is already updated locally, or `async_request_refresh()` for buffered updates instead of immediate full refreshes Example of batching entity creation: ```python # Instead of multiple calls: if param_sensors: async_add_entities(param_sensors) if status_sensors: async_add_entities(status_sensors) if alarm_sensors: async_add_entities(alarm_sensors) # Collect all entities and add at once: all_entities = [] all_entities.extend(param_sensors) all_entities.extend(status_sensors) all_entities.extend(alarm_sensors) if all_entities: async_add_entities(all_entities) ``` This approach reduces function call overhead, improves code clarity, and can significantly improve performance when dealing with large numbers of operations.