COVID-19 didn't invent remote work, hybrid teams, or flexible hours. It removed the excuse not to try them. That's the actual lesson — not "the future of work changed forever," which is the line every consultancy deck used, but something narrower and more useful: the pandemic proved that most of the operational objections leaders had been making for a decade — "we can't trust people at home," "culture only happens in a building," "clients expect us in the office" — were preferences dressed up as constraints.
I've coached leaders through the whole arc: the March 2020 scramble, the 2021 hybrid experiments, the 2022-23 return-to-office tug of war, and the quieter settling-in that's happened since. What I've watched, close up, is which leaders adapted and which ones just waited for permission to go back to how things were. The difference wasn't industry, company size, or even how digital the business already was. It was whether the leader treated the disruption as a diagnostic — a chance to see what was actually load-bearing in how they ran their team — or as an interruption to be endured until normal resumed.
That's the frame I want to give you here, because it's the one that's still useful years later. This isn't a recap of what changed — you lived it, you don't need reminding. What's useful now is the filter for deciding, in each new disruption, whether you're protecting something real or just protecting habit.
The trust test COVID-19 ran on every manager
Remote work in 2020 was a forced trust experiment, and the uncomfortable version goes like this: a lot of managers failed it privately while passing it publicly. Output didn't collapse. Teams that were given autonomy largely kept delivering. But plenty of managers spent the period installing more monitoring, not less — activity trackers, mandatory camera-on policies, check-in cadences that existed to manufacture the feeling of control rather than to solve an actual delivery problem. If your instinct under uncertainty is to watch people harder rather than clarify outcomes and get out of the way, that's not a remote-work problem. That's a trust deficit that remote work made visible. It was there before 2020; the office just hid it, because presence looked like productivity even when it wasn't.
I wrote about this pattern in more depth here — the point that matters most is that trust erosion is rarely announced. People don't tell you they've stopped believing you'll back them under pressure. They just stop volunteering the honest version of what's happening, and you find out months later, expensively.
My evaluation lens for any "future of work" claim
Every year since 2020 someone declares a new verdict on hybrid, RTO, or the "death of the office." Most of these claims are unfalsifiable — vibes dressed as trend analysis. When a client asks me whether they should follow a workplace trend, including the ones in this article five years ago, this is the lens I actually use.
Five questions before you copy anyone's workplace policy
- What decision does this actually change?: If a trend piece doesn't tell you to do something different on Monday morning, it's commentary, not guidance. Discard it.
- Whose retention curve improved, not just whose survey scores did?: Engagement surveys move with novelty. Watch twelve-month attrition and internal mobility instead — those numbers don't lie to please a facilitator.
- Would this policy survive a bad quarter?: Flexible-work perks that get quietly withdrawn the moment revenue dips were never a culture commitment. They were a hiring-market concession. Know which one you're building before you announce it.
- Does it require me to trust people, or to watch them?: Any future-of-work initiative that increases surveillance while claiming to increase flexibility is solving the wrong problem — usually a management confidence problem, not a productivity one.
- Can my worst manager execute this without a rulebook?: If a policy only works when delivered by your best people manager, it's not a system — it's a hope. Design for the median manager, not the exception.
That fourth question is the one I go back to most. Nearly every RTO mandate I've seen argued for in boardrooms since 2022 was, underneath the stated rationale about collaboration or culture, actually about a small number of senior leaders who didn't trust their middle managers to run distributed teams. The mandate solved the trust problem by removing the conditions that exposed it. It didn't build the management capability that would have let flexibility work. That's expensive avoidance, not a workplace strategy.
Where employee experience and customer experience actually connect
There's a causal claim I'll defend without hedging: an organisation cannot sustainably deliver good customer experience on the back of a demoralised or distrusted workforce. Not because it's a nice HR sentiment, but because service quality is a downstream output of discretionary effort, and discretionary effort is the first thing people withdraw when they feel unheard. You cannot script your way around that. Scripts handle the transaction; they don't handle the moment a customer needs someone to actually care, and caring isn't something you can mandate from someone who's already checked out.
This is why I push leaders to treat employee listening mechanisms — real ones, not annual engagement surveys that get actioned in a slide nobody revisits — as a leading indicator for customer churn, not a separate HR initiative running in parallel to the "real" business metrics. If your best people are quietly disengaging, your customer experience is already degrading; you just haven't seen it show up in the numbers yet, because that lag is usually two to three quarters.
Automation and AI didn't replace the case for human judgement — they raised the bar for it
The other pandemic-era prediction that needs revisiting: that AI and automation would offload enough routine work to free people up for higher-value judgement calls. That part was directionally right. What most predictions missed is that this shift raises the skill floor for everyone, not just specialists. When the routine, low-judgement work gets automated, what's left is disproportionately the ambiguous, high-stakes, relationship-dependent work — the calls that used to get made by whoever happened to be free. That's harder work, not easier work, and most organisations haven't upgraded their training or their hiring criteria to reflect it. They automated the queue and left the people who used to triage it undertrained for what's left.
If you're rolling out AI tools into a team right now, the leadership question isn't "will this save time." It almost always will, on paper. Ask a harder one instead: are you developing your people's judgement fast enough to handle what the automation leaves behind? Teams that get this backwards end up with faster processes and worse decisions, because nobody built the capability for the harder work that's now theirs.
Why the office-versus-remote argument was always the wrong argument
I want to be direct about something I got asked constantly between 2021 and 2023: which side was right, the remote-first camp or the return-to-office camp? My answer disappointed most people who asked, because it wasn't a side. Both camps were arguing about location as if location were the causal variable, when location is almost always a proxy for something else — usually trust, sometimes cost control, occasionally a genuine collaboration need that got wildly overstated to justify a decision the leadership team had already made for other reasons. Once you strip the location argument away, what's left is a much smaller and more honest set of questions: does this specific team, doing this specific work, need to be physically co-located to do it well? For some teams — early-stage product teams wrestling with genuine ambiguity, teams onboarding a wave of junior hires who need to absorb tacit knowledge by osmosis — the answer is often yes, and I'll say so plainly. For most teams doing execution work against a known brief, the answer is no, and dressing the no up as a yes to satisfy a real-estate lease or a leader's discomfort with not seeing people is where trust gets spent.
The organisations that navigated this well didn't run a survey and average the preferences. They asked the harder question team by team, sometimes role by role, and accepted that the honest answer would look inconsistent from a policy-memo perspective. Inconsistency that's grounded in the actual nature of the work is a feature. Uniformity imposed for administrative convenience is the thing that erodes trust fastest, because people can tell the difference between a decision made for them and a decision made about them.
The mistake I still see leaders make when a disruption hits
Here's a pattern I've watched repeat itself past COVID-19, into the 2022 downturn, into the AI adoption wave that followed — leaders treat disruption as a problem to be managed back to baseline rather than a signal about what baseline should have been all along. The instinct is understandable. Uncertainty is uncomfortable, and "get back to normal" is a genuinely soothing goal. But normal was never neutral. It was a set of choices someone made, under a previous set of constraints, that calcified into "how we do things" long after the original justification stopped applying. Every disruption is an invitation to re-examine which of your practices are load-bearing and which ones just never got questioned. Most leaders decline the invitation, because re-examining is slower and riskier than reverting, and reverting looks like decisiveness in the short term even when it's the opposite.
The leaders I respect most from this period weren't the ones who called the remote-versus-office debate correctly. They were the ones who used the forced disruption as licence to ask questions about their organisation that they'd been avoiding for years — questions about which meetings actually needed everyone in them, which approval chains existed because someone once got burned rather than because they added value, which managers had been coasting on presence rather than earning trust through delivery. That audit was available to every leader in 2020. Most of them did it under duress for about six months and then quietly stopped, the moment it became socially acceptable to stop. The few who kept doing it as a standing discipline are the ones running noticeably better organisations now.
Five actions I actually recommend, in order
- Name the purpose before you name the policy — Don't lead with "three days in the office" or "fully remote." Lead with what the organisation is for, then let the policy follow. Policies without a stated purpose get treated as arbitrary — and get resisted accordingly.
- Build flexibility around outcomes, not optics — Define what "done well" looks like for a role, then let people choose how they get there. If you can't define the outcome, that's the actual gap — not the location of the desk.
- Invest in your middle managers before you invest in tools — Most flexible-work failures are management-capability failures wearing a technology costume. A manager who can't run a good one-to-one in person won't magically run one well over video.
- Cut the process that only existed to signal presence — Audit your meetings, your status updates, your check-ins. Ask which ones exist to move work forward and which exist to reassure someone that work is happening. Kill the second category.
- Decide what you won't compromise on before the next disruption hits — You will face another forcing event — a downturn, a technology shift, a talent-market swing. Decide now which of your current practices are genuine values and which were pandemic-era conveniences, so you're not improvising your principles under pressure next time.
The distinction I want you to leave with
If you take one thing from this, take this: "future of work" was never really about where people sit. It was a stress test of how much leaders actually trusted the people they'd hired. COVID-19 didn't create a new set of workplace values — it exposed which values organisations already had and which ones were theatre. Some leaders discovered their teams were more capable and more trustworthy than five years of management orthodoxy had assumed. Others discovered how thin their trust actually was the moment they lost the ability to watch people work in person.
I don't think the interesting question, this far out, is remote versus hybrid versus office. That debate has mostly settled into whatever equilibrium suits each business's actual constraints, and further argument about it is mostly noise. The interesting question is whether your organisation used the disruption to build genuine management capability — the kind that survives ambiguity, that doesn't need physical presence as a proxy for trust — or whether it just waited out the disruption and reverted the moment it was socially acceptable to do so.
You can tell which one happened by looking at what survived. Genuine capability shows up in how your managers handle the next disruption, whatever it turns out to be — a downturn, a new technology, a talent shortage. Theatre shows up as a policy document nobody follows and a return-to-office memo written by someone who never explained why presence mattered for the work in question. The pandemic didn't hand any of us a new leadership model. It handed us a mirror. What you do with what you saw in it is still the whole job.
Further reading: 10 Essential Characteristics for Successful Leaders, 10 Cross-Functional Leadership Skills Every Manager Needs, 5 Creative Managerial Strategies to Solve Poor Performance
Further reading: What Is Cross-Functional Team Alignment?, Handling Leadership Stress Without Losing Efficiency