In Anatomy of the Unsought Finding. Serendipity: Origin, History, Domains, Traditions, Appearances, Patterns and Programmability Pek Van Andel writes (as one conclusion) the following.

In general the role of serendipity in science, technology and art is underestimated. This is mainly and unintentionally caused by the way we rationalize a posteriori about theoretical and experimental research and its results, when we publish. The not strictly rational, chronological or searched components (like chance, fortuitous, accidental, surprising, unsought, (n)ever dreamt of, unknown, etc.), which have led to these results are therefore underestimated and sometimes even banned from the theater and totally hidden behind the decor. The next step is that pure rationality becomes the norm, not only regarding the results, but also regarding everything that has led to these results. Scientists then report their results as following directly and logically from their initial hypothesis, omitting possibly crucial (pseudo)serendipitous events. Reading and interpreting such articles as 'the inside story' about the discovery, can unwillingly brainwash the researcher in such a way and to such an extent that he neglects during his own research the flowers along the road that can form a nicer bouquet than those he is looking for. This can cause a loss of serendipity: the aim and/or plan spoils the journey. A successful researcher or manager has one open eye for sought findings and another open eye for unsought findings.