Two companies. Same hybrid policy: three days in office, two days remote. Radically different outcomes.
Company A saw productivity climb 15%, with engineering satisfaction scores reaching five-year highs. Company B saw attrition spike, productivity drop, and engineers describing the policy as "the worst of both worlds."
The policy was identical. The implementation was everything.
I saw this contrast play out last year when both companies brought me in—one to understand their success, one to diagnose their failure.
Engineers came in Tuesday through Thursday because that's when everyone came in. But "everyone" wasn't actually everyone—team members in different time zones, those with exceptions, new hires who'd been promised fully remote. So the in-office days meant a lot of people in the office, but not the specific people any given engineer needed to collaborate with. The spontaneous collaboration that justified the policy rarely happened.
The remote days weren't true remote work either. Engineers had to stay near enough to commute on office days, couldn't take advantage of remote flexibility for travel or relocation. They had all the constraints of being local without the benefits of being in-person.
"We got the costs of both approaches and the benefits of neither," the VP of Engineering told me. "People are frustrated, attrition is up, and I'm not sure what we're getting for it."
Hybrid work can work—many companies are making it work—but it requires thoughtful design rather than a simple policy that splits the difference between remote and in-person. The naive approach assumes that alternating between office and home gives you collaboration when you need it and focus when you need it. Reality is messier: you get collaboration overhead all the time and focused collaboration rarely.
At SmithSpektrum, I've advised dozens of companies on hybrid work design[^1]. The ones that succeed treat hybrid as its own model that needs its own structures, not as a compromise between two other models. The ones that struggle try to preserve office-centric work patterns while adding some remote—and get something that doesn't work as either.
What Hybrid Often Gets Wrong
The problems with naive hybrid implementation are predictable.
Mandated days create compliance rather than collaboration. When everyone comes in on Tuesday because the policy says Tuesday, you get bodies in the building. You don't get productive collaboration—that requires the right people working on the right things, not just physical proximity. Mandatory presence breeds resentment without producing benefits.
Mixed-presence meetings are terrible. A meeting with some people in a conference room and others on video creates information asymmetry. The people in the room can see each other's reactions, have side conversations, read body language. The remote participants miss all of this. Nobody is happy. Either go all-remote for meetings or all-in-person—mixed is a bad experience for everyone.
Collaboration doesn't happen spontaneously just because people are in the same building. The "bumping into people at the coffee machine" benefit requires the right people to bump into each other, which doesn't happen reliably. Intentional collaboration—planned discussions about specific work—isn't helped by physical proximity if it's not designed into the schedule.
Deep work gets sacrificed on both ends. In the office, the interruptions and ambient noise make focused work hard. At home, the mental model of "office days for collaboration" leads engineers to feel they should be available and responsive, preventing deep focus there too.
Commuting costs remain but are distributed differently. Engineers still need to live near the office. They still spend time and money commuting—just on fewer days. For some, this is worse: commuting three days a week feels like all the pain with none of the routine.
Principles for Effective Hybrid
Hybrid can work, but it requires different design principles than either fully remote or fully in-office.
Be intentional about what happens where. Don't let "wherever you happen to be" determine what work gets done. Explicitly decide: these types of work benefit from in-person presence; these types work better remote. Then design your practices around those decisions.
| Hybrid Model | Office Days | Best For | Watch Out For |
|---|---|---|---|
| Flexible hybrid | 0-5 (employee choice) | Autonomy-driven cultures | Coordination challenges |
| Structured hybrid | 2-3 fixed days | Teams needing sync time | Creates two-tier culture |
| Team-determined | Varies by team | Different team needs | Inconsistent experience |
| Anchor days | 1 fixed + flexible | Guaranteed overlap | Still sub-optimal for some |
| Remote-first + gatherings | 0 regular + quarterly | Distributed teams | Gathering logistics |
In-person time should be for things that genuinely benefit from presence: relationship building, complex collaborative discussions, brainstorming sessions, workshops, celebrations, difficult conversations. If you're doing these things over video anyway, in-person days aren't adding value.
Remote time should be for things that benefit from focus and flexibility: deep technical work, asynchronous collaboration, documentation, individual problem-solving. If engineers spend their remote days in back-to-back video calls, they're not getting the benefit of remote.
Make meetings work in one mode or the other. If a meeting is important enough to have, everyone should be in the same mode—either all in-person or all remote. Mixed-mode meetings should be the exception, not the norm. This might mean scheduling important meetings on days when the relevant people are all in the office, or defaulting to video even when some people are co-located.
Synchronize presence when it matters. The value of in-person presence is multiplied when the right people are present together. A team meeting in person with everyone is valuable. A team meeting with half the team remote isn't particularly better than all remote. Coordinating who's in the office when turns random presence into useful presence.
Team-Level Design
Hybrid works better when designed at the team level rather than imposed uniformly.
Different teams have different collaboration needs. A team building a tightly integrated feature might benefit from intensive in-person collaboration. A team handling bugs and maintenance might need less synchronous coordination. A team spanning multiple time zones might not be able to be in-person together at all.
Let teams design their own hybrid patterns. Give them the autonomy to decide what works: which days to be in the office together, what activities to prioritize for in-person time, how to handle people who can't make office days. Teams know their work better than a company-wide policy can account for.
Team agreements should be explicit. When is the team expected to be in the office together? What happens on those days versus remote days? How do we handle meetings when some people can't make office days? Making these agreements explicit prevents misunderstandings and resentment.
Some teams may work better fully remote. If a team's work doesn't benefit from in-person collaboration, forcing them into the office is pure cost with no benefit. Allow teams to opt out of hybrid if their work patterns support it—especially teams that are already distributed across locations.
Scheduling That Works
The schedule matters more in hybrid than in either pure model.
Clustered office days work better than scattered ones. If the team agrees to be in the office Tuesday through Thursday, you get three days of genuine in-person collaboration. If individuals scatter their three days across the week, you get more total office-days but less actual overlap—people coming in to find that their collaborators aren't there.
Weekly rather than daily thinking helps. Instead of "Tuesday is an office day," think "this week we have two in-person collaboration sessions planned, so let's all be there for those." The days become a function of the work rather than an arbitrary pattern.
Protect focus time on remote days. If remote days fill up with meetings, they're not serving their purpose. Establish team norms around meeting-free blocks, expectations for responsiveness, and when it's acceptable to decline meetings.
Plan in-person time deliberately. Don't leave office days to chance—decide what you're going to do with the face-to-face time. Planning sessions, design discussions, one-on-ones, team lunches. If there's no plan, the benefit of being in-person doesn't materialize.
The Role of the Office
What the office is for changes in hybrid work.
The office becomes a collaboration hub rather than the default work location. Optimize it for the types of work that happen there: meeting rooms rather than rows of individual desks, spaces for whiteboarding and discussion, comfortable areas for informal conversation.
Individual workstations matter less. If engineers have good home setups, they don't need assigned desks that sit empty half the week. Hot-desking or flexible seating can work—though some people strongly prefer having their own space.
Amenities can pull people in voluntarily. Free lunch, good coffee, comfortable environment, collaboration tools—these make the office appealing rather than just mandated. When engineers want to come in because the experience is good, you get genuine benefit. When they come in because they have to, you get compliance.
The office as social connector has value beyond specific work activities. The casual conversations, the serendipitous encounters, the shared meals—these build relationships that support remote collaboration later. This value is real but hard to mandate.
Managing Hybrid Effectively
Managing hybrid teams requires adapting management practices.
Communication needs to be more explicit. In an office, managers get ambient information—they see who's working, who's frustrated, who's having trouble. In hybrid, this ambient information exists on some days and not others. Managers need to actively seek out information rather than absorbing it passively.
One-on-ones matter more. The relationship-building that happens informally in an office needs dedicated time in hybrid. More frequent, more intentional one-on-ones help maintain connection.
Documentation becomes critical. If decisions get made in in-person conversations, remote workers are excluded. If information only spreads verbally, it doesn't reach people who weren't there. Writing things down—decisions, context, rationale—makes hybrid work accessible to everyone.
Performance management must be output-focused. You can't evaluate based on presence when presence is partial and variable. What did the person accomplish? Did they meet their commitments? These questions work regardless of where someone was when they did the work.
Include remote participants intentionally. When planning in-person activities, remember those who can't attend. What do they miss? How do you bring them in? The worst hybrid cultures create two tiers: in-person insiders who have information and relationships, and remote outsiders who don't.
When Hybrid Doesn't Work
Hybrid isn't right for every situation.
Highly collaborative, tightly-integrated work may benefit from consistent in-person presence. If the work requires constant real-time coordination, hybrid creates friction. Some teams should be in-person by design.
Highly independent, asynchronous work may not need any in-person presence. If engineers work independently on well-defined tasks with clear async coordination, mandatory office days are overhead that doesn't contribute.
Distributed teams can't be hybrid. If team members are in different cities or countries, in-person presence isn't an option—you're remote-first by necessity. Pretending otherwise creates problems.
Engineers who strongly prefer one mode over the other may not thrive in hybrid. Some people love the office; some people love working from home. Forcing everyone into a middle ground that's nobody's preference creates dissatisfaction.
If hybrid isn't working, consider whether you should be hybrid at all. Sometimes the answer is fully remote; sometimes it's fully in-person for certain teams. Hybrid for the sake of compromise, when neither aspect is working, is worse than committing to one model.
The company whose hybrid policy was the worst of both worlds? They redesigned their approach.
Instead of mandated days, they moved to team-determined patterns. Some teams chose to be in the office together two days a week; others chose to be fully remote with occasional in-person gatherings. They redesigned meeting norms—all-video or all-in-person, no mixed mode. They invested in making office days valuable: scheduled collaboration sessions, team lunches, designated brainstorming time.
Satisfaction improved. Attrition dropped. The people who came in came in for reasons that made sense. The people who stayed remote stayed remote because their work didn't benefit from presence.
"We stopped trying to find a universal answer," the VP reflected. "Hybrid works when it's designed around the work—not when it's a policy that tries to please everyone and pleases no one."
References
[^1]: SmithSpektrum remote work advisory, hybrid model design, 2021-2026. [^2]: Microsoft Work Trend Index, "Hybrid Work Research," 2025. [^3]: Gallup, "The Advantages and Challenges of Hybrid Work," 2024. [^4]: Harvard Business Review, "Making Hybrid Work Work," 2023.
Designing a hybrid work model? Contact SmithSpektrum for distributed team strategy and organizational design.
Author: Irvan Smith, Founder & Managing Director at SmithSpektrum