The return to office debate keeps getting framed as a choice between two camps. Leaders who want people back. Employees who want to stay home. This week, two stories pulled that framing apart, and they did it from opposite directions.

One was a CEO who used to count cars in the parking lot and stopped. The other was a Wharton researcher saying the idea that remote work is obviously good isn’t true. Read together, they point at the same thing. The office was never really the point. It was the proof.

And when the proof stops working, leaders have to decide what they were actually trying to prove.

Key takeaways

  • Presence in the office is a signal. The standards underneath it are reliability, helpfulness and relationships that let work move.
  • New research from Wharton and UT Austin found remote work can erode the informal help networks that new hires depend on most.
  • Counting cars in a parking lot measures attendance, not contribution. Mandates built on that signal read as distrust.
  • The better move is to define and reward the work that presence used to stand in for, such as helping colleagues and mentoring, wherever people sit.

What’s happening in the return to office debate

On October 5, Fortune reported on a Wharton and UT Austin study of roughly 760 employees at a multinational financial services company in the United States, the United Kingdom and India. Wharton management professor Peter Cappelli and UT Austin communications professor Y. Jasmine Wu drew on 38 group interviews conducted in late 2023. Their paper, “How Remote Work Is Reshaping the Office,” was published online in July in California Management Review.

Their core finding is about what they call relational capital: knowing whom to ask for help, feeling comfortable asking, and actually getting it. Sending a message was easy. Getting a reply was not. People focused on their own metrics and pushed colleague requests to the next day. Employees with established relationships got help quickly. Those without struggled, and one new hire spent three months trying to get an answer. The researchers also note that employees highly valued the ability to work from home, and that the company’s hybrid rollout had its own problems: reduced office space, shared desks that kept teammates apart, and attendance rules that were not consistently enforced.

Their recommendations are worth noting. Coordinate office days so teams are together. Plan structured collective activities. Put helping colleagues and mentoring into formal job descriptions. Count teamwork in performance evaluations and promotions, not just individual metrics.

Two days earlier, Fortune profiled World Wide Technology CEO Jim Kavanaugh, who described once looking at cars in the parking lot after 6 p.m. on Fridays. During the pandemic he asked himself whether he would mandate that all employees be in the office 100% of the time, and decided not to. The company now uses a hybrid approach where flexibility depends on the role and business unit. “We treat our people with a level of trust, like adults,” he told Fortune. George Mason University’s Kevin Rockmann put the other side plainly in the same piece: “When you’re forcing people to come in, you don’t trust them, and employees are sensitive to that.”

And on October 1, HR Dive covered a Resume.io analysis of the cost of returning to the office. Based on an August survey of 1,000 full-time U.S. employees, Resume.io estimated the base cost at $6,736 a year, rising to $16,538 with childcare, plus the equivalent of 23 eight-hour workdays lost to commuting. As Resume.io career expert Amanda Augustine put it, “Employers may not be able to eliminate the RTO tax, but acknowledging that it exists matters.”

What this reveals: presence was a signal, not the standard

In Chapter 13 of The Next Turn, “Building a Shared Definition of Good Work,” I separate standards from signals. Standards are what the organization actually needs: quality, reliability, accountability, sound judgment, follow-through, care, contribution. Signals are the visible cues we use to infer those things: time, presence, responsiveness, visible effort.

Reliability is the principle. Sitting at a desk for a fixed number of hours was one signal.

The Next Turn

The parking lot was a signal. A full lot on Friday evening told a leader something felt right. It did not tell him who had helped a struggling colleague, who had made a sound call, or who had quietly kept a project from going sideways.

Here is what I find useful about the Wharton study. It does not defend the signal. It defends a standard. Helping a colleague. Answering the new hire. Mentoring. That work used to happen almost by accident because people were in the same room. The structure of work held it up without anyone having to name it. When the structure changed, the work did not disappear because people stopped caring. It disappeared because no one had ever defined it, measured it or rewarded it.

That is the collision I describe in Chapter 2. The definition of work, meaning what people believe work is for and what it should give back, has shifted toward flexibility and autonomy. The structure of work, meaning how work actually flows from person to person, has shifted to messages, async replies and fewer shared rooms. Focus only on what employees want and you lose the help network. Focus only on attendance and you send the message Rockmann describes.

There is also a value exchange here. In Chapter 5, I argue that people now assess value first and build loyalty only if the exchange proves itself. Whatever you think of a career site’s cost estimate, employees are doing that math. If people are asked to absorb real costs to come in, the days in the office have to deliver something that a video call cannot. If those days are spent on Zoom at a shared desk away from the team, the exchange stops making sense.

So the question is not office or home. The question is what the office was supposed to produce, and whether you have defined that clearly enough to get it either way.

Your next turn

  • Name the standard behind your policy. Write one sentence on what in-person time is supposed to produce on your team. If the answer is “people being seen,” that is a signal, not a standard.
  • Make helping count. Following the Wharton recommendations, add helping colleagues and mentoring to role expectations and to how performance and promotion are discussed.
  • Design the in-office days. Coordinate them so the team is actually together, and spend them on work that benefits from being in one room.
  • Set a response norm for new hires. Agree on how quickly questions from newer colleagues get an answer, and who they go to first.
  • Ask your team about the exchange. What does an office day give them that a remote day does not? What does it cost them? Listen before you mandate.

Frequently asked questions

Is return to office better for productivity than remote work?

The evidence in this week’s coverage does not settle that. The Wharton and UT Austin study found remote work weakened the informal help networks people rely on, especially new hires, while employees highly valued working from home. The useful question for leaders is which specific outcomes in-person time is meant to produce.

How do I build trust with a hybrid team without monitoring attendance?

Define what reliability looks like in your team’s actual work: response norms, follow-through, how decisions get explained, how colleagues get help. Then evaluate against those standards. Attendance tells you where someone was. It does not tell you what they contributed.

Why do new employees struggle more in remote and hybrid setups?

In the Wharton study, people with established relationships got help quickly while those without struggled, and one new hire waited three months for an answer. Newer employees have the least relational capital, so they depend most on deliberate structures such as coordinated office days, clear contacts and response expectations.

The question underneath the policy

I don’t think leaders pushing for more office time are wrong to worry. Something real does get lost when no one answers the new hire for three months. And I don’t think employees who value flexibility are wrong either. Both sides are pointing at something true.

The work is to stop arguing about the signal and start defining the standard. Once you know what the office is for, you can design for it, in the building and out of it.

If you are working through this with your own leadership team, there is more on the two turns and the shift from signals to standards on the The Next Turn home page.

So here is the question I’d sit with this week: if every car disappeared from your parking lot tomorrow, how would you know who was doing the work that matters?

Sources