Office attendance set a post-pandemic record in 2026, with Kastle’s 10-city index averaging 56.3% and Tuesday alone hitting 66.0%. Both are records only within the post-2020 era, and the shape of the week is more revealing than the level: midweek attendance runs roughly double Monday and Friday, and 73% of organizations name Tuesday as their busiest day. On the policy side the market has converged on three required days, now the rule at 66% of structured-hybrid companies, up from 53% in mid-2024. The figures below come from Kastle Systems badge data, Flex Index policy tracking, and CBRE research.
TL;DR
- Kastle’s 10-city Back to Work Barometer hit a record 56.3% weekly average (Kastle Systems)
- That was roughly half a point above the previous record (Kastle Systems)
- The single-day Tuesday reading reached 66.0%, a post-pandemic record (Kastle Systems)
- Top-grade buildings reached 95.5% on Tuesdays (Kastle Systems)
- 66% of structured-hybrid companies now require three office days (Flex Index)
- That is up from 53% in mid-2024 (Flex Index)
- The average required in-office commitment is near 2.78 days per week (Flex Index)
- Roughly two-thirds of Fortune 100 companies run some hybrid policy (Flex Index)
- Just under a third of Fortune 100 companies require full-time attendance (Flex Index)
- 73% of organizations report Tuesday as their highest-attendance day (CBRE)
- Midweek occupancy runs roughly double Monday and Friday (Kastle Systems)
- The three-in, two-remote pattern is now the near-standard setup (Flex Index)
- Records here are post-pandemic records, not a return to pre-2020 levels (Kastle Systems)
1. The Badge Data
Observed attendance finally set records in 2026 after four years of plateau. Kastle’s 10-city Back to Work Barometer averaged 56.3% for the record week, about half a point above the prior high, with a single-day Tuesday reading of 66.0%. Both numbers describe badge swipes at buildings Kastle secures, which is a real sample of real behaviour rather than a survey of intentions.
| Metric | Value | Source |
|---|---|---|
| 10-city weekly average, record week | 56.3% | Kastle Systems |
| Margin above previous record | approx. 0.5 points | Kastle Systems |
| Single-day Tuesday reading | 66.0% | Kastle Systems |
| Status of that Tuesday figure | post-pandemic record | Kastle Systems |
| Trophy-grade A+ buildings on Tuesday | 95.5% | Kastle Systems |
| Cities in the index | 10 | Kastle Systems |
| Measurement basis | badge swipes | Kastle Systems |
| Comparison to pre-2020 baseline | still far below | Kastle Systems |
The gap between the 10-city average of 56.3% and trophy buildings at 95.5% is the most under-reported number here: the recovery is concentrated in the best buildings, not spread across the market. Source: Kastle Back to Work Barometer hits all-time post-pandemic highs.
2. Policy Converged on Three Days
Employers spent four years arguing about the right number and have now largely settled on the same one. 66% of companies running structured hybrid require three office days, up from 53% in mid-2024, and the average required commitment across that population sits near 2.78 days per week. The three-in, two-remote pattern is the near-standard setup rather than one option among many.
| Metric | Value | Source |
|---|---|---|
| Structured-hybrid companies requiring three days | 66% | Flex Index |
| Same figure in mid-2024 | 53% | Flex Index |
| Change over roughly eighteen months | up 13 points | Derived from Flex Index figures |
| Average required in-office days | approx. 2.78 per week | Flex Index |
| Dominant pattern | three in, two remote | Flex Index |
| Fortune 100 running some hybrid policy | roughly two-thirds | Flex Index |
| Fortune 100 requiring full-time attendance | just under a third | Flex Index |
| Most common Fortune 100 rule | three days | Flex Index |
The direction of travel is toward three, not five: even among the largest employers, full-time in-office remains a minority policy at just under a third. The 13-point move from 53% to 66% in roughly eighteen months is also worth reading carefully, because it reflects two separate flows rather than one. Some companies tightened from two days to three. Others loosened from four or five days to three. Convergence on the same number from both directions is what a market settling on a norm looks like, and it is a better indicator of where policy lands than any single company’s announcement. Source: Flex Index reports.
3. Tuesday Runs the Week
No finding in this field is more consistent across independent datasets than the shape of the week. 73% of organizations report Tuesday as their highest-attendance day, and midweek occupancy runs roughly double Monday and Friday. That distribution is what a three-day mandate looks like when employees choose which three days, and it produces facilities that are overcrowded on Tuesday and near-empty on Friday.
| Metric | Value | Source |
|---|---|---|
| Organizations reporting Tuesday as peak day | 73% | CBRE |
| Midweek occupancy versus Monday and Friday | roughly double | Kastle Systems |
| Busiest days overall | Tuesday, Wednesday, Thursday | Kastle Systems |
| Peak single-day reading | 66.0%, Tuesday | Kastle Systems |
| Weekly average for comparison | 56.3% | Kastle Systems |
| Gap between peak day and weekly average | approx. 10 points | Derived from Kastle figures |
| Trophy-building Tuesday reading | 95.5% | Kastle Systems |
| Required days under the dominant policy | 3 | Flex Index |
A 10-point spread between the peak day and the weekly average means capacity planned for Tuesday sits substantially idle the rest of the week, which is the operational cost of letting employees pick their days. Employers face a genuine trade-off here rather than a simple failure of enforcement. Mandating which days people attend would flatten the curve and improve space utilisation, but it removes the flexibility that makes hybrid tolerable to employees and is the part most cited in retention surveys. Leaving the choice open preserves that flexibility and produces buildings sized for a Tuesday that happens once a week. Most large employers have chosen the second, which is why the midweek peak has persisted for four years rather than resolving. Source: Kastle on the emerging hybrid work pattern.
4. Records Without Recovery
The word “record” is doing heavy lifting in most coverage of these numbers. A 56.3% weekly average is a post-pandemic record and remains far below the pre-2020 baseline, against which Kastle’s index is constructed. Attendance has improved steadily but has not returned, and the weekly distribution is structurally different from what preceded it.
| Metric | Value | Source |
|---|---|---|
| Record weekly average | 56.3% | Kastle Systems |
| What that record is relative to | the post-pandemic era only | Kastle Systems |
| Comparison to pre-2020 | still far below | Kastle Systems |
| Direction across recent quarters | gradual improvement | Kastle Systems |
| Pace of that improvement | slow and steady, not step change | Kastle Systems |
| Shape of the week versus pre-2020 | far more uneven | Kastle Systems |
| Average required office days | approx. 2.78 | Flex Index |
| Implied maximum attendance under that policy | approx. 56% of a five-day week | Derived from Flex Index figures |
That last derived row is the most useful reconciliation in the dataset: an average requirement of 2.78 days out of five is about 56%, which is almost exactly the observed 56.3% weekly average. Policy and behaviour have converged. Workplace sentiment context sits in our workplace burnout statistics. Source: Kastle peak day data.
5. Reading the Two Datasets Together
Kastle and Flex Index get quoted interchangeably and measure fundamentally different things. Kastle observes badge swipes at buildings it secures; Flex Index records stated company policy across a large employer sample. Policy and attendance diverge routinely, and the interesting analysis lives in the gap between them rather than in either number alone.
| Metric | Value | Source |
|---|---|---|
| What Kastle measures | observed badge swipes | Kastle Systems |
| What Flex Index measures | stated company policy | Flex Index |
| Kastle sample basis | buildings Kastle provides security for | Kastle Systems |
| Kastle city coverage | 10 cities | Kastle Systems |
| Flex Index coverage | large multi-employer policy set | Flex Index |
| Observed weekly average | 56.3% | Kastle Systems |
| Policy-implied weekly rate | approx. 56% | Derived from Flex Index figures |
| Peak-day observed reading | 66.0% | Kastle Systems |
| Organizations naming Tuesday as peak | 73% | CBRE |
The near-match between observed 56.3% and policy-implied 56% is genuinely striking, but it is a coincidence of aggregates: within it, some employees exceed their requirement and others ignore it entirely. Broader work-pattern context sits in our AI in the workplace statistics and tech layoffs statistics. Source: Kastle Getting America Back to Work.
Summary: Return to Office by the Numbers
| Metric | Value | Source |
|---|---|---|
| 10-city weekly average, record | 56.3% | Kastle Systems |
| Margin above previous record | approx. 0.5 points | Kastle Systems |
| Peak single-day Tuesday reading | 66.0% | Kastle Systems |
| Trophy-grade buildings on Tuesday | 95.5% | Kastle Systems |
| Midweek versus Monday and Friday | roughly double | Kastle Systems |
| Cities in the Kastle index | 10 | Kastle Systems |
| Structured-hybrid companies requiring three days | 66% | Flex Index |
| Same figure in mid-2024 | 53% | Flex Index |
| Change over eighteen months | up 13 points | Derived |
| Average required in-office days | approx. 2.78 | Flex Index |
| Dominant pattern | three in, two remote | Flex Index |
| Fortune 100 with a hybrid policy | roughly two-thirds | Flex Index |
| Fortune 100 requiring full-time attendance | just under a third | Flex Index |
| Organizations naming Tuesday as peak day | 73% | CBRE |
| Gap between peak day and weekly average | approx. 10 points | Derived |
| Policy-implied weekly attendance rate | approx. 56% | Derived |
| Comparison to pre-2020 baseline | still far below | Kastle Systems |
| What Kastle measures | badge swipes | Kastle Systems |
| What Flex Index measures | stated policy | Flex Index |
Methodology and Sources
- Occupancy readings, peak-day data, building-grade breakdowns, and the shape of the week come from Kastle Systems’ Back to Work Barometer, which aggregates badge access data from buildings Kastle provides security services for across 10 US cities (record highs, peak day data, hybrid work pattern, Getting America Back to Work, occupancy by day of week).
- Policy requirements, required-day averages, and Fortune 100 breakdowns come from Flex Index, which tracks stated flexible-work policies across a large employer sample (Flex Index reports, Flex Index stats).
- Day-of-week attendance patterns are corroborated by CBRE research as reported in industry coverage (Propmodo office occupancy trends).
- Data watch: Kastle’s barometer is not a measure of the US office market. It covers 10 cities and only buildings where Kastle provides access control, a sample that skews toward larger commercial properties, and it is indexed against a pre-2020 baseline rather than expressing true seat utilisation. Records reported here are post-pandemic records only. Flex Index measures policy as stated by employers, not compliance, and a company can require three days without enforcing them. The near-match between the policy-implied 56% and the observed 56.3% is an aggregate coincidence and does not imply individual-level compliance. Rows marked as derived are arithmetic on published figures.
- Last updated: August 2, 2026. We update this roundup quarterly as Kastle and Flex Index publish new readings.