Context: Systematic literature reviews have become common in software engineering in the last decade, but challenges remain. Goal: Given the challenges, the objective is to describe improvement areas in writing primary studies, and hence provide a good basis for researchers aiming at synthesizing research evidence in a specific area. Method: The results presented are based on a literature review with respect to synthesis of research results in software engineering with a particular focus on empirical software engineering. The literature review is complemented and exemplified with experiences from conducting systematic literature reviews and working with research methodologies in empirical software engineering. Results: The paper presents three areas where improvements are needed to become more successful in synthesizing empirical evidence. These three areas are: terminology, paper content and reviewing. Conclusion: It is concluded that it must be possible to improve the primary studies, but it requires that researchers start having synthesis in mind when writing their research papers.