@[youtube](https://youtu.be/n7of0iXGZ-Y) **Talk summary** Several scientific fields are experiencing a reproducibility crisis, in which hypothesis-driven studies are failing to replicate. Poor reproducibility has been linked to several factors, but one of the most pertinent issues is analytical flexibility. There are thousands of ways to analyse a typical dataset, but many articles only report the methods that yield statistically significant results, which support the reported hypotheses. Reporting hypotheses that are subject to flexible analysis as “pre-specified” without time-stamped evidence reduces study credibility, as these results are less likely to replicate. While the issue of analytical flexibility is becoming well-known for primary research, it has received less attention for meta-analysis. In this talk, I present two remedies for analytical flexibility in meta-analysis: pre-registration and multiverse analysis. Reducing analytical flexibility, in parallel with releasing raw data and analysis scripts, will improve the credibility of meta-analysis.