User feedback on software usage is utilised by developers to improve their software, and software product forums are platforms rich in software-related user feedback. However, previous studies have mainly focused on analysing forum user feedback as individual sentences, which can lead to missing insights and a lacking understanding of the overall context of forum posts. This work examines software forum user feedback in forum posts to investigate the differences between sentence and post-level content analysis. We manually evaluated software forum posts collected from two open-sourced software forums and discovered five new types of user feedback that can only be identified in the post-level analysis. Additionally, we examined the association between sentence classifications found within software forums. Our results indicate that contextual information complimenting product improvement insights can be found in software forums. This information can be used to reduce the manual effort required to chase up missing contextual information when working through an issue. We also provide insights into the progression of posts in software product forums at the thread level. Our findings reveal the importance of looking at forum-based user feedback at a post level to identify new insights on user feedback for software improvements. The file contains the post-level classification dataset created from this study as well as our coding guideline used for the process.