Why 'I'll get back to you tomorrow' stopped being an acceptable answer

Nobody sent a memo announcing that replying within the hour was now the norm. There was no vote, no policy update, no HR announcement titled “new expectations for response time.” It happened the way most real culture shifts happen — gradually, through a thousand small examples of what got rewarded and what got quietly noticed — and by 2026 it’s fully in place. The evidence for it isn’t a survey asking people how they feel. It’s telemetry showing what they actually do.

The old shape of a workday

For most of the office era, a workday had one shape: a single block, roughly 9 to 6, inside which replying was expected, and outside which it mostly wasn’t. An email sent at 7pm reasonably wouldn’t get a reply until the next morning, and nobody read anything into that gap — the building was closed, more or less literally. The email itself was already the async-friendly medium of that era: slower than a phone call by design, and everyone understood the implicit SLA as “sometime tomorrow, probably,” not “before I finish this coffee.”

That single-block shape mapped cleanly onto a single physical constraint: everyone left the same building around the same time, so “after hours” was a shared, visible, unambiguous fact. Nobody had to infer it.

The shape it has now

Remove the shared building and the shared clock goes with it. Microsoft’s own 2025 telemetry — measured from actual product usage across the Microsoft 365 install base, not a self-reported survey — shows what replaced the single block: three separate peaks across the day. Forty percent of people already online before 6am are triaging an overflowing inbox before their calendar’s first meeting. A midday peak runs through 117 emails and 153 chat messages a day, on average. And a third peak starts after dinner: meetings starting after 8pm are up 16% year over year, more than 50 messages a day now go out or come in outside core business hours, and nearly a third of active workers are back checking their inbox by 10pm.

Why the expectation tightened instead of loosening

The intuitive prediction, going into remote and hybrid work, was that async by default would relax response-time pressure — no one’s watching you leave your desk, so why would anyone expect an instant reply? What actually happened is closer to the opposite, and the mechanism is worth naming: without a shared building acting as the default cutoff, the cutoff has to be negotiated implicitly, thread by thread, and the default that fills a vacuum is usually the more anxious one. If nobody has explicitly said “after 6pm is fine to wait,” the safer assumption for anyone worried about how responsive they look is to answer now. Multiply that instinct across a whole company simultaneously reachable at all hours, and you get exactly the three-peak pattern in the data — not because any single person decided to work more, but because nobody opted out first.

This is also why disconnection policies, on their own, tend to underperform as a fix: a written policy is one more explicit signal competing against dozens of implicit ones — a manager who answers Slack at 9pm, a peer who replies to email on Sunday, a culture where the fast responder gets remembered as “on top of it.” The policy says one thing; the pattern everyone actually observes says another, and people calibrate to the pattern.

The timezone amplifier

Distributed teams add a second force pushing in the same direction: 30% of meetings now span multiple time zones, per the same 2025 telemetry, up sharply from the pre-remote era when a meeting invite assumed everyone shared a building and a clock. Every cross-timezone meeting creates at least one participant for whom “normal hours” and “the meeting’s hours” don’t match, and that person’s inbox behavior starts leaking into everyone else’s expectations — if a colleague in a nine-hour-offset office regularly replies at what looks like 11pm their time, “always available” quietly becomes the visible norm for the whole thread, even for people who could have waited until morning. A single distributed team member effectively resets what “reasonable” looks like for the group, without anyone deciding that on purpose.

What this means for anyone sending email into that inbox

If you’re the one sending into this environment rather than managing it, the practical read is less about ethics and more about arithmetic: the average B2B inbox now absorbs 120-plus sales emails a week on top of the internal volume above, arriving into an inbox that’s being triaged in short, anxious bursts rather than read calmly once a day. Decision fatigue research suggests position in that queue matters — email 40 gets judged differently than email 4, even holding the content constant — and a three-peak day means there are now three separate “email 4” moments instead of one, each a slightly better shot than landing mid-scroll during the noisy midday peak. It’s part of why the data on best send times keeps favoring mornings even as the exact best hour stays contested: mornings are still closer to a fresh, low-fatigue read than the parts of the day this new shape has fragmented.

Frequently asked questions

Is it true that remote workers are actually working more hours than office workers did?+

The clearest 2025-2026 telemetry doesn't cleanly separate 'more total hours' from 'the same hours spread across more, shorter sessions.' What it does show clearly is a change in shape: three distinct daily peaks (early morning, midday, and after 8pm) instead of one continuous block, which is a different problem than simple overwork even when total hours are similar.

Do disconnection policies actually fix the always-on pressure?+

On their own, not reliably — a written policy is competing against every implicit signal employees see daily, like a manager replying at 9pm or a peer answering email on a weekend, and people tend to calibrate to the observed pattern rather than the written rule.

What does this mean for when to send a cold email?+

It reinforces morning-leaning send windows, but for a more specific reason than 'mornings are best': a fragmented workday now creates several distinct fresh-inbox moments instead of one, and the goal is landing near the start of one of those triage bursts rather than in the noisy middle of it.

Send into the fragmentation, don’t fight it

You can’t reset how your prospects’ companies structure their day, but you can control how your own sending adapts to it. Norbelys’s sending schedule respects the sender’s and the recipient’s working hours instead of blasting a list the instant a campaign launches, and its unibox routes every reply back to one place so a response that lands during someone’s 6am triage burst doesn’t sit unnoticed until you happen to check later. If your outreach is still timed around your own convenience rather than the actual rhythm your prospects’ inboxes now run on, that’s the fastest fix available before you touch a single subject line. See how Norbelys’s plans work and start sending on a schedule built for the workday people actually have now, not the one that stopped existing.