Designing an analysis to size remote work's effect on productivity after COVID-19

Has COVID-19, together with the shift to working from home that came with it, changed how productive the workforce is? If so, by how much? Walk through how you would answer this. ```hint Separate the causes The pandemic changed demand, hiring, health and childcare in the same months that it changed where people worked. Decide what you will count as output and as input, then how you would tell the effect of working from home apart from everything else that moved at the same time. ``` ### Clarifying Questions - Whose productivity: one company's employees (ours, for instance), a particular industry, or the economy as a whole? - Which period: the lockdown months when nearly everyone was remote, the hybrid arrangements that followed, or both? - What decision should this inform, such as an office-attendance policy, or is the aim to understand what happened? - Can I use internal company data, such as work output and HR records, or only public statistics? ### What a Strong Answer Covers - A chosen scope and a definition of productivity as output relative to input, with measures that fit each kind of work - Data sources, a pre-pandemic baseline, and the ways the measures can shift for reasons unrelated to output - A method for isolating the effect of working from home from the pandemic's other shocks, and the assumption it rests on - An estimate stated as a range and broken down by type of work, with an honest level of confidence - Effects beyond short-term output, such as attrition, hiring reach, onboarding and collaboration, and how they bear on any decision ### Follow-up Questions - Leadership wants one number to justify an office-attendance mandate. What would you give them, and with which caveats? - How would you measure productivity for work whose output is hard to count, such as research or design? - If output looked flat but attrition fell, how would you weigh the two?