Online Social Networks (OSN) have been increasingly used as sources of information for different applications, ranging from business, politics, and public services. However, there is a lack of information on OSN platforms' behavior that may impact big data processing and real-time services. In this paper, two of the most widely used social networks, Instagram and Twitter, are investigated to broaden the understanding of how each platform's message characteristics influence data completeness and latency. We performed a series of experiments to emulate data posting and collection automatically. Our results increase the level of transparency of the platforms' internal behavior, showing that both can deliver data with reasonably low latencies and high completeness, but Twitter can be up to eight times faster when it comes to multimedia messages.