dsh-ab-sample

The most overlooked step in A/B testing is sample-size estimation. Teams often launch an experiment only to discover, weeks later, that the statistical power was insufficient, having wasted real traffic and development cycles. dsh-ab-sample moves that check to the very beginning of the workflow: pass in significance level, target power, baseline conversion rate, and expected lift, and receive a precisely computed sample size along with a detailed breakdown of every calculation step. The plugin is written in plain Node.js with zero network calls, so results are returned instantly without spinning up any external service. For data analysts, product managers, and engineers who are in the experiment-design phase, it acts as a local calculator you can invoke right inside the conversation—no copy-pasting into a web tool, no context switching—so you can focus on the analysis and decisions that actually move the product forward.