Citation-based metrics are important in determining careers, so it is unsurprising that recent publications advise prospective authors on how to write highly cited papers. While such publications offer excellent advice on structuring and presenting manuscripts, there are significant downsides, including: restrictions in the topics researched, incentives to misconduct and possible detriments to motivation, innovation and collegiality. Guides to writing highly cited papers also assume that all citations are equal, ignoring new directions in bibliometric research identifying "quality" and perfunctory citations. Rather than pursuing citations, with the uncertainty about their significance and the potential negative consequences, authors may fare better by following evidence from several disciplines indicating that persistence, a focused research program, good methodology and publishing in relevant journals are more important in career development and disciplinary influence than the odd star paper. Research administrators could encourage such steps by considering innovative new multivariate assessments of research productivity, including assessing social impact.