A people-operations lead walks the floor on the first mandated Monday back and counts fourteen people at desks built for sixty. The welcome-back lunch was catered, the banner is still taped to the wall, and the office is mostly empty. Above her, a CEO decided the team returns. Below her sits a team that does not want to. She cannot change the policy, and she is accountable for whether the office feels worth the commute.
That gap, between the mandate and the morale, is where this article lives. A manager who wrote to the workplace advice column Ask a Manager in January 2025 named it precisely: “stuck in the middle,” no authority over the rule, full responsibility for the room. Whether the title is HR director, people-operations lead, internal communications, or facilities manager, the job is the same.
Search “return to office ideas” and the first page hands over a hundred activities: welcome-back bingo, scavenger hunts, branded merchandise, happy hours. The ideas that actually rebuild culture share a trait those lists skip. They run on a schedule, and someone measures whether they worked. A welcome-back lunch fills the office for a day. Rebuilding the social fabric that frays when a team stops sharing a room takes a repeatable program. What follows sorts the ideas by what evidence says employees show up for, explains the mechanism that turns a good day into a weekly habit, and gives an operator a way to tell whether any of it worked.
Why most return-to-office ideas fade by the third week
The welcome-back event is built for day one. The problem operators actually face arrives in week six. By then the banner is down, the catered lunch is a memory, and the commute is exactly as long as it ever was. Attendance drifts back toward whatever people were doing before the event. The day was optimized; the weeks were not.
What Mandates Miss
Culture does not erode the way a machine breaks, all at once and visibly. It thins. Shared context, knowing what other teams are working on, who just joined, what the running joke is, degrades slowly when people stop overlapping in person. Weak-tie relationships, the loose connections with people outside the immediate team, go first, because they were never held together by scheduled meetings. They lived in hallways and coffee lines. A two-hour event cannot rebuild months of thinned connection, the same way one big dinner does not repair a year of not calling a friend.
The numbers say the mandate alone is not closing the gap. MIT Sloan Management Review reported in November 2024 that US office utilization had been “hovering at 50% of pre-pandemic norms” for a year, even as Amazon, Google, IBM and other large employers issued high-profile return mandates. Policy moved in one direction. Behavior mostly did not follow.
So attendance is won week by week, not declared once. The ideas that rebuild culture have to be a cadence, not a stunt.
What employees actually show up for (and what employers keep buying instead)
A return-to-office budget gets allocated the way most budgets do, toward things that are easy to point at. A bike room. A monthly happy hour. Branded merchandise on every desk. The harder question, the one the spending tends to skip, is whether any of it removes an actual reason people stay home.
The recruitment firm Hays put that question to both sides. In research reported by Raconteur in September 2023, covering mostly UK workplaces, the gap between what employers offered and what workers wanted was wide and consistent. Bike storage had been installed by 49% of employers; 13% of workers said it would bring them back. More social events were on offer from 48% of employers; 27% of workers wanted them. The incentives workers actually asked for ran the other way. Subsidized or free food was wanted by 45% of workers and offered by only 32% of employers. Subsidized travel was wanted by 46% and offered by just 18%.
The pattern is plain
The pattern is plain. Workers want the direct cost of showing up covered: the commute, the lunch, the expenses that make the office a worse financial deal than the kitchen table. Employers tend to buy amenities, the bike room and the social calendar, things that look like effort and photograph well but do not touch what the day costs an employee.
A separate US survey by NORC at the University of Chicago, fielded with an SHRM panel and published in November 2023, found the blunter version of the same problem: nine in ten employers had introduced no new return-to-office incentive at all. When NORC asked employees what would help, the leading answer was additional pay for in-office work, followed by food, amenities, and commuter benefits. The list is mostly about cost.
The misconception worth correcting outright is that a better building does the job. HR Dive, reporting on an EY survey of more than 17,000 employees, noted that workers in “Class A” real estate showed no significant difference in preferred working location compared with employees in ordinary offices. A nicer lobby does not change the arithmetic of a ninety-minute round trip. Before an operator spends, the test is simple: does this perk remove a real barrier, or does it just look like care?
Return-to-office ideas that earn the commute
Picture the planning meeting: a whiteboard with twenty-five activity ideas and a budget that covers maybe six of them. The useful question is not which ideas are good. It is which ones do a job the others cannot. Sorted by the job they do, the ideas worth running fall into four groups, and a program needs all four.
Anchor a shared day
Pick one weekday the whole team is in, and make that day worth the trip. Two cases reported by Raconteur show what a strong anchor day does to attendance. Time Etc, a UK virtual-assistant company, introduced free chef-cooked lunches twice a week; according to its founder and chief executive, attendance rose by roughly 20% on those days. The Stag Company, a Brighton events business, added social events, buffet lunches, and free drinks on Thursdays; its head of marketing reported that office attendance doubled, with the average employee coming in once a week rather than once a fortnight. Both are company-reported figures rather than audited numbers, but the direction agrees: concentrate the good things on one named day and people organize around it.
Rebuild weak ties
The connections that fray fastest are the loose ones: the colleague two teams over, the new hire nobody outside their own pod has met. Those weak ties carry information and a sense of belonging across an organization, and they are the first thing to thin when a team stops sharing a building. Programming that deliberately crosses team lines does repair work a department standup never will, whether that is a cross-department mixer, an interest-based group, a lunch-and-learn, or a session that pairs new hires with long-tenured staff.
Mark milestones in the room
Recognition, project-win celebrations, and work anniversaries quietly migrated into chat channels over the remote years. Moving the emotionally meaningful moments back on-site, a closed deal announced to the floor, a five-year anniversary marked in person, gives people a reason to be there that a recurring calendar invite cannot.
Remove the friction
The barriers the data identified get addressed directly here: workspace customization budgets, commute or transit support, reliable quiet space for focused work, and predictable desk booking so nobody drives in to hunt for a seat. The strongest evidence that friction removal works comes from a randomized controlled trial published in Nature in June 2024. At the Chinese travel company Trip.com, a hybrid schedule cut employee attrition by about a third, with especially large retention gains among staff who had long commutes, and no measurable drop in performance over the following two years of reviews. It is a single company, but a controlled trial rather than a survey, which makes it the firmest causal evidence in this space: taking the commute away on some days kept people without costing output.
The anchor-day mechanism: why one great day beats three scattered ones
A common mandate reads “three days a week in the office” and stops there. It names a quantity and skips the coordination. Left to choose their own days, people scatter, and the predictable result is a floor that is half-empty every day of the week. An employee who drove in expecting energy and collaboration finds the same empty desks waiting at home, and quietly concludes the office is not worth the trip. The mandate succeeded on paper. The culture goal failed.
The anchor day fixes a coordination problem, not a motivation problem. When everyone knows Tuesday is the day, the floor is reliably full on Tuesday. The lunch is worth catering, leadership is present, and cross-team work can happen because the other team is there too. Employees left to self-organize already do a rough version of this. Office attendance clusters hard around mid-week: Kastle Systems, which measures badge swipes across thousands of US office buildings, finds Tuesday the most consistent peak and Monday and Friday the troughs.
The operator’s move is to stop fighting that pattern and lean into it. Spreading a thin budget evenly across five days produces five mediocre days. Putting the catered lunch, the leadership presence, the all-hands energy, and the cross-team work on the single day the office is already fullest produces one day people plan around. Connection needs a critical mass of people in the same place at the same time, and only the anchor day reliably delivers it. One excellent anchored day rebuilds more than three thinly attended ones. Pick the day, defend it on the calendar, and make it the day nobody wants to miss.
Turn return-to-office moments into proof the culture is rebuilding
Suppose the anchor day works. The lunch lands, two teams that had not spoken in months end up sketching on a whiteboard, a project win gets a real round of applause. Then everyone goes home, and the part of the team that was out that day has no idea any of it happened. The good day evaporates instead of compounding.
Culture change is hard to manage because it is mostly invisible. A manager cannot point to it on a dashboard the way they point to revenue. Events that produce something visible, photos, a short written recap, a recognition post, a running record of in-office moments, solve two problems at once.
Momentum
The first is momentum. People who skipped the anchor day and then see what they missed are more likely to show up next time. This is the healthy version of FOMO: not manufactured scarcity, just an honest record of a good day doing its own recruiting.
Signal
The second is signal. A steady stream of real in-office moments is evidence, for the team and for the leadership that approved the budget, that the program is doing something. When the only trace of a return-to-office program is a line item, it reads as a cost. When it leaves a visible trail, it reads as a culture.
The practical versions are vendor-neutral and cheap to start: a recap channel where someone posts a few photos and a paragraph after each anchor day, a shared photo stream from events, a screen or a physical wall in the office showing the quarter’s moments, content made by employees at the event rather than staged afterward by marketing. For a recurring anchor day or a milestone celebration, a self-running photo station carries the same job without leaning on a designated photographer: Simple Booth’s HALO kit sets up in under a minute and lets employees send themselves each shot by QR code, email, or text, with every photo landing in one shared gallery the recap channel can draw from. The principle is to design return-to-office moments to leave a trail. A moment that leaves no record only ever happens once.
How to tell if it is working: a return-to-office scorecard
Six months in, a finance lead asks the people team a fair question: is the program working? If the only honest answer is “the lunches seem popular,” the budget is already in trouble. A return-to-office program is a budget line, and it should be defended with numbers. Four are worth tracking.
Attendance lift
The first is attendance lift: occupancy on programmed days measured against ordinary baseline days. The second is sentiment, captured by a short pulse survey that repeats often enough to show a trend. The third is the ratio of voluntary to mandated attendance, which matters because compliance and presence are not the same thing. Owl Labs’ State of Hybrid Work found that 43% of hybrid workers already “coffee badge,” showing up just long enough to be seen before leaving. Badge data that only counts entries will miss this; a program rebuilding culture wants people who stay. The fourth is retention, especially regrettable attrition among the people the company least wants to lose.
The arithmetic is worth making explicit
The arithmetic is worth making explicit. Take a 150-person company running a weekly anchor day with a catered lunch at $18 a head. That is $2,700 per anchor day, roughly $135,000 a year across about fifty working weeks. Add a modest budget for activities, recognition, and the occasional larger event, and the program lands near $150,000 a year.
Set that against the cost of losing people. Gallup estimates that replacing an employee costs roughly one-half to two times their annual salary, counting recruiting, lost productivity, and ramp time. For a mid-level employee earning $65,000, that is between about $32,500 and $130,000 per departure. A $150,000 program that prevents even two such departures a year has more than paid for itself when replacement costs land in the upper half of that range; at the low end, it needs to retain four or five people to break even. Either way, the comparison an operator brings to finance is not “lunch costs $135,000.” It is “the program costs $150,000, and a single avoided resignation recovers a fifth to nearly all of it.”
The exact figures will differ by company. The point is that a program defended with attendance deltas, sentiment trends, and a retention comparison survives a budget review. One defended with “people seemed to enjoy it” does not.
Mistakes that quietly stall a return-to-office program
A program rarely fails loudly. Attendance just never climbs, the pulse survey flatlines, and at the next budget review someone proposes cutting it. The post-mortem usually finds one of these.
- Mandating without a reason. A return policy delivered as an order, with no explanation of what the office is actually for, reliably breeds resentment. MIT Sloan Management Review reported in 2024 that blanket mandates tend to damage engagement and raise attrition among exactly the high performers a company least wants to lose. Employers who gave at least a quarter’s notice and explained the reasoning had an easier time, according to HR Dive’s reporting.
- One-and-done events. A single welcome-back with nothing scheduled behind it. The empty week-six office is the predictable result.
- Copying another company’s perk list. Lina Tonk, a marketing executive at HR-software firm isolved, told HR Dive that companies “are just trying to mimic each other” instead of “looking at themselves and at their business goals.” A perk that worked elsewhere may not touch the barrier a given team faces.
- Running with no measurement. Without attendance and sentiment data, the program cannot be improved or defended, and it dies at the first budget review.
- Treating it as HR’s project alone. If leadership mandates the office and then works from home, the team notices immediately. The anchor day needs the people who set the policy in the room.
- Ignoring the commute math. Owl Labs estimates in-office workers spend about $15 a day getting to work and $18 on lunch. A program that never acknowledges that cost is asking employees to subsidize their own return.
Return-to-office ideas are easy to find and easy to copy. The hard part, the part the listicles skip, is running them as a measured, repeatable program built around one reliably good day. The manager stuck in the middle cannot rewrite the mandate. She can make Tuesday the day nobody wants to miss, give the good moments a way to travel, and walk into the next budget meeting with a retention number instead of a hopeful adjective. That is the difference between decorating a return and rebuilding a culture.
Sources
- Ask a Manager (2025). “Our CEO is demanding we return to the office but people don’t want to, and I’m a manager stuck in the middle.” https://www.askamanager.org/2025/01/our-ceo-is-demanding-we-return-to-the-office-but-people-dont-want-to-and-im-a-manager-stuck-in-the-middle.html
- MIT Sloan Management Review (2024). “Return-to-Office Mandates: How to Lose Your Best Performers.” https://sloanreview.mit.edu/article/return-to-office-mandates-how-to-lose-your-best-performers/
- Raconteur (2023). “Which incentives will actually get your employees back in the office?” https://www.raconteur.net/talent-culture/return-office-incentives
- NORC at the University of Chicago (2023). “Employees Want Compensation for the Costs of Coming into the Office, While Few Employers Have Introduced Incentives to Return.” https://www.norc.org/research/library/workers-employers-diverge-return-office-incentives-compensation.html
- HR Dive (2023). “4 mistakes to avoid when calling employees back to the office.” https://www.hrdive.com/news/return-to-office-policy-hr-mistakes/695326/
- Nature (2024). “Hybrid working from home improves retention without damaging performance.” https://www.nature.com/articles/s41586-024-07500-2
- Kastle Systems (2026). “Getting America Back to Work: Occupancy by Day of Week.” https://www.kastle.com/safety-wellness/getting-america-back-to-work-occupancy-by-day-of-week/
- Owl Labs (2025). “State of Hybrid Work 2025.” https://www.owllabs.com/state-of-hybrid-work/2025
- Gallup (2019). “This Fixable Problem Costs U.S. Businesses $1 Trillion.” https://www.gallup.com/workplace/247391/fixable-problem-costs-businesses-trillion.aspx