In recent years, the use of big data in economics, political science and other related disciplines has become increasingly more common. Against this backdrop, the construction of ever larger datasets necessitates a more nuanced focus on the improvement of the quality of research design through the use of experimental and quasi-experimental methods that drove the credibility revolution in empirical economics. However, the shift towards bigger datasets may been seen as a necessary but insufficient condition to establish causal relationships of interest. In the absence of rigorous empirical scrutiny, bigger datasets may provide a misleading and noisy insights that may hinder better understanding of the various phenomena and distort normative implications for policymakers. Recent advances in empirical methods and data collection allow for the use and construction of better data and offer numerous opportunities and challenges in unravelling theoretically insightful and policy-relevant effects from empirical research design. Critiques argue that in pursuing clean and plausible empirical strategies of identification, the pendulum of design-based studies has sought good answers rather than asked good questions. Through a brief digression into political economy, economic growth and law and economics, these concerns and potential trade-offs are discussed, showing that there is little reason for audacious concerns.