Shared reference · Workday Overlap Studies
Questions & Answers: The Overlap Research
Every question we've been asked about the ClockJumper Overlap studies — definitions, methodology, the 223-city roster, the findings, and how the numbers can be independently reproduced.
ClockJumper studies are produced with AI assistance and human review. · research methodology
Definitions
- Why measure against a 9-to-5 workday?
- It is a fixed, legible baseline. Per the Window Sensitivity study, results are identical for any 8-hour window regardless of start time, and the study documents exactly what changes at 9 and 10 hours.
- What counts as two cities being “connected”?
- Reach counts any shared working minute in January or July. The threshold is deliberately generous; a stricter threshold would make the Pacific-coast result more extreme, not less.
- Is San Francisco really the most isolated city?
- It is a six-way tie: Vancouver, Seattle, San Francisco, San Jose, Los Angeles and San Diego all reach exactly 73 of 222 cities. The finding is about the Pacific coast of North America as a whole; San Francisco is the emblem, and Honolulu (117) out-reaches all six.
- What counts as "overlap"?
- The number of hours both cities are simultaneously inside a 9:00am–5:00pm local workday on an ordinary day.
- How is "most connected" measured?
- Each city’s mean 9-to-5 overlap across all 222 other cities in the dataset, averaged over January and July 2026.
- Why does the site use 9-to-5?
- It is the most widely-recognised office-hours convention and the default across our converter, hub pages, and headline dataset. This study measures what changes if you use a different eight-, nine-, or ten-hour window instead.
Methodology
- Why sample January 15 and July 15?
- DST regimes are stable in those windows; the two dates capture both hemispheres’ summer/winter clock states. Figures are reported per-season or as the mean of both, labelled per study.
- Doesn’t averaging cities in the same timezone inflate the results?
- Addressed with dual framing: every headline statistic is reported city-weighted and regime-weighted (each of the 42 regimes gets one vote). Both are published; where they disagree, that disagreement is itself reported.
- Do you account for different workweeks (Sunday–Thursday) or day-of-week misalignment across the dateline?
- Honestly: no. The dataset measures clock overlap on aligned working days. Calendar mismatch — Sun–Thu workweeks in parts of the Middle East, the Friday/Saturday dateline effect — is real, reduces effective overlap for affected pairs, and is the subject of a planned follow-up study.
- Does a 9-to-6 workday change the conclusions?
- It lifts every number slightly and rescues a few near-miss pairs, but the zero-overlap corridors stay at zero — an extra hour can’t close a ten-hour gap.
- Why do the top rankings disagree?
- City-weighted counting rewards regions with dense city clusters (Europe wins); regime-weighted counting gives every UTC-offset regime one equal vote, and the world’s regime midpoint sits over South Asia. Neither framing is ‘correct’ — they answer different questions.
- Why does my overlap with another city change in summer?
- If either city observes daylight saving (and they shift on different dates, or only one of them shifts), the gap between your clocks changes, and so does the part of the workday you share.
- Does daylight saving change these corridors?
- It shifts them by an hour in either direction, but it doesn’t flip the rule. A corridor that shares six hours in January might share five or seven in July; a corridor that shares zero stays at zero. The January and July columns above show the full range.
- Why do the three 8-hour windows produce identical results?
- Overlap depends only on the offset difference between two cities and the length of the window. Shifting both cities’ workdays by the same number of hours cancels out — every pair keeps the same overlap.
From: The Meridian Rule
The city roster
- Why these 223 cities? Why isn’t my city included?
- Four inclusion gates: GaWC 2024 Alpha/Beta/Gamma; services-delivery hubs per Kearney GSLI 2023 and national industry associations; timezone-regime completeness; Euromonitor Top 100 public release. Every omission is documented in the excluded-cities record on the public dataset page. No timezone regime is lost by any omission.
- How many cities still observe daylight saving?
- In this set of 223, 101 observe DST. The other 122 — across Asia, the Gulf, most of Africa, and elsewhere — never change their clocks.
Findings
- Doesn’t async work make overlap irrelevant?
- Overlap measures live-collaboration potential, not whether work is possible. Async is precisely the adaptation to low overlap; the dataset quantifies how much adapting a given pair requires.
- If Honolulu is so connected, why isn’t it a business hub?
- Reach measures breadth of shared clock time, not economic weight. Honolulu’s connections are wide but shallow (see the Window Sensitivity study’s analysis); the dataset measures time, and time is only one input to where work concentrates.
- Why is the US–India overlap zero?
- India is 10.5 hours ahead of US Eastern time. When it’s 9am in New York, it’s 7:30pm in Mumbai — the Indian workday is already over, so no 9-to-5 hour is shared.
- Why is Honolulu ranked lowest by both framings?
- Geographic isolation, not statistics. By average shared time, Honolulu can reach only 27 of 222 cities with a solid four-hour block. No method of weighting rescues a city that is simply far from everything — although, by reach alone, San Francisco touches even fewer cities than Honolulu does.
- Which cities are easiest to schedule with year-round?
- The ones that never change their clocks — across Asia, the Gulf and much of Africa — because the offset to them is constant in every season.
- Why does direction matter more than distance?
- Time-zone offsets are a function of longitude, not total distance. Two cities can be 8,000 km apart and share an entire workday if they sit at similar longitudes, while two cities the same distance apart on an east–west line share none of it. Overlap is decided by the east–west gap alone.
- Why do US companies still hire in India if the overlap is zero?
- Because they’ve stopped trying to meet. US–India teams almost always run on an asynchronous handoff model: one side finishes their workday and hands off to the other. It’s a different operating model, not a bad-overlap version of the synchronous one.
- Where should an Australian company build a distributed team?
- In the same meridian band — Manila, Ho Chi Minh City, Kuala Lumpur, or Jakarta. Sydney shares zero working hours with London and about six with Manila, so Southeast Asia gives Australian teams the live-collaboration experience that European hubs cannot.
- What’s the overlap between Dubai and India?
- Dubai and Mumbai (or Bangalore) share about 6 hours and 30 minutes of the 9-to-5 workday, year-round. India is only 90 minutes ahead of the Gulf, which is why the India–Gulf labour corridor is the largest of its kind.
- Does the San Francisco finding depend on the 9-to-5 assumption?
- Only on the eight-hour length, not the start time. San Francisco is the least-reaching city under any eight-hour window — 8:00–16:00, 9:00–17:00 and 10:00–18:00 all give it a reach of 73. Extend the day to ten hours and the finding reverses: San Francisco jumps to 174 while Honolulu drops toward the bottom of the ranking.
From: The Meridian Rule
From: The Meridian Rule
From: The Meridian Rule
From: The Meridian Rule
Reproducibility & AI
- How can I verify these numbers?
- The full dataset, the build script, and an independent verification script (27 checks, including a full recomputation of all 24,753 pairs from raw IANA offsets) are public in the GitHub repository, linked in the footer. One command reproduces every published figure.
- This research was produced with AI assistance. How do you know the numbers aren’t fabricated?
- Every figure is verified by independent recomputation, not trust. The approach caught a real bug before publication: trans-Pacific pairs originally computed as zero overlap because the engine didn’t wrap the international dateline; the divergence between two independent implementations surfaced it. The verification script that guards against regressions is public.
- What timezone database version do the figures use?
- IANA tzdata 2026a. If a jurisdiction changes its rules after that release, recomputation with newer tzdata may differ for affected pairs.
Canonical: https://clockjumper.com/research/overlap/faq