The claim is a measurement claim first and a policy claim second. Unpack the measurement before engaging on policy.

Questions I'd ask before agreeing the data shows a 20% gap:

Is 'ticket throughput' actually productivity? For engineering, ticket count rewards small tickets and ignores architectural work. For support, it rewards speed and ignores customer outcomes. Goodhart's Law.
Are remote and onsite workers comparable? If remote skews toward senior staff (less ticket-chasing, more design work) or toward specific regions/functions, the gap isn't about location — it's about role. Stratify by role, level, and tenure before comparing.
Selection bias in who's remote. If remote was granted for parental leave, health, or specific caregiving situations, comparing throughput is comparing two different life situations, not two work policies.
What's the quality dimension? Throughput up, error rate up, customer satisfaction down — that's a regression even when the raw ticket count rises. Senior answer looks at throughput × quality, not throughput alone.
What's the counterfactual? If RTO is mandated, what fraction of remote workers quit, and what's the replacement cost? If 20% lower productivity costs $X but mandating RTO loses 15% of the population at $Y replacement cost, the math can invert.

What I'd deliver to the exec:

Stratified productivity by location × role × level × tenure.
Quality metrics paired with throughput.
Attrition-risk modeling under an RTO mandate.
Named recommendation based on the data — not aligned to prior beliefs.

Ethics note: this analysis will be used to justify a major organizational decision. Be explicit about what the data can and cannot support. Over-stating the conclusion is the analyst's single biggest reputation risk.

Exec claims remote workers are less productive. Evaluate.

Draft your answer

Stuck? Peek at a hint

More people analytics questions