The reply lands at 11:47 pm. A team member posted a question in Slack twenty minutes earlier, half-expecting to handle it themselves in the morning, and the founder answered before midnight, fully, with a decision attached. Everyone reads it. Nobody says anything. But the lesson registers anyway: the founder is always there, the founder will answer, and the safest move when you are unsure is to wait for the founder. The always-available founder thinks they are removing friction. They are installing it, just somewhere they cannot see, in the place where their team would otherwise have learned to decide.
How availability became a signal
Availability turned into a virtue because it is the cheapest one to display. You cannot easily show the team that you are thinking hard about strategy, but you can show them you answered an email at 6 am. The work itself has obliged the performance. Microsoft’s Work Trend Index special report on the infinite workday, published June 17, 2025 and drawn from trillions of anonymized Microsoft 365 signals, found that 40% of people online at 6 am are already reviewing email for the day’s priorities, that the average worker receives 117 emails a day and 153 Teams messages every weekday, and that meeting volume in the evening has climbed. The day now starts before you are out of bed and does not end at a clean hour. For a founder, being reachable across that whole expanse is not heroic. It is just what the tools make easy to perform, and the team reads the performance as the standard.
The trouble is that the signal is read by people who depend on you, and they adjust their behavior to it. A founder who answers everything immediately is not modeling diligence. They are advertising that the fast path to an answer runs through them, which is an invitation the team will accept.
The decision-making cost
This is the cost that gets missed, because it does not show up as exhaustion. It shows up as a quiet centralization of judgment. Shannon Barrett, who coaches founders on exactly this, calls it the Founder Bottleneck: the structural condition where the core intelligence of the business, its decision logic and standards and judgment, exists only in the founder’s head, so delegation fails not because the hire is incapable but because they lack the context to decide the way the founder would. When the founder is always available to supply that context on demand, the team never has to internalize it. They just ask. The asking feels efficient to everyone in the moment and compounds into dependence.
The symptoms are recognizable once named. Cloud Nine Business Coaching’s breakdown of the founder bottleneck, from January 2026, lists decision latency, where teams wait days for sign-off on small items and momentum dies; a bus factor of one, where critical knowledge and relationships live only in the founder’s head; learned helplessness, where strong hires stop taking initiative because they expect to be overruled; and a revenue plateau, often in the $1 million to $10 million range, where more effort produces less growth. The report makes the timing point that matters: centralized decision-making is a genuine feature early, at Series A or before, because the founder’s judgment is a fast heuristic for messy calls. It becomes the constraint when the team grows from 5 to 50 and the decision surface area explodes while the operating model does not. Always-available is the behavior that keeps the early-stage model running long after the company has outgrown it.
The always-available founder thinks they are removing friction. They are installing it in the place where their team would otherwise have learned to decide.
The output cost
Then there is the founder’s own work, which is supposed to be the thing the availability protects but is in fact the first casualty. Cal Newport, the Georgetown computer-science professor whose 2016 book Deep Work defined the term, draws a hard line between deep work, distraction-free concentration that creates new value and is hard to replicate, and shallow work, the logistical tasks done while distracted. Constant availability is a machine for converting the first into the second. The founder who keeps Slack open to answer in twenty minutes is, by definition, never more than twenty minutes into a single thought.
The cost of that is not linear, which is the part people underestimate. Sophie Leroy’s research, published in Organization Science in 2009 under the title “Why Is It So Hard to Do My Work?”, identified what she named attention residue: when you switch from one task to another, a residue of your attention stays stuck on the first, and it impairs your performance on the second, especially when the first task was left unfinished. The interruption does not cost you the two minutes of the reply. It costs you the degraded thinking for some stretch afterward. Gloria Mark’s work at UC Irvine, summarized in her 2023 book Attention Span, puts a number on the recovery: an average of 23 minutes to return to the original depth of attention after an interruption. A founder who fields ten such interruptions in an afternoon has not lost ten short intervals. They have lost the afternoon. Newport’s own illustration of the alternative is Donald Knuth, the computer scientist who has had no email address since January 1, 1990, on the stated grounds that his role is “to be on the bottom of things” and that what he does “takes long hours of studying and uninterruptible concentration.” Most founders cannot live like Knuth. The point is the direction, not the extremity.
The relationship and the body
The availability does not stay at work either. The same Microsoft data that maps the 6 am inbox check maps an evening that fills with meetings and messages, which means the device that holds the team also sits at the dinner table. There is no clean separation between the founder who answers Slack at 11:47 pm and the partner or child in the room while they do it. The 23-minute recovery cost applies to home too: the reply that takes a minute leaves residue on the hour that follows. Organizational researchers have given this its own name, the spillover-crossover model, and the empirical work on after-hours work-related ICT use finds that the emotional exhaustion it generates does not stay with the worker. It crosses over to the spouse, so the cost of the midnight reply is paid twice, once by the founder and once by the person sitting next to them. The founder is present in the building and absent in the chair.
The exit ramp
The way out is not a productivity hack. It is a deliberate reduction in your own availability, sold to the team as a gift rather than a withdrawal. The mechanics are unglamorous and well understood. Batch communication into windows instead of answering continuously, the practice the attention-residue research directly supports. Externalize the invisible operating system Barrett describes, writing down the decision standards and guardrails so the team can apply them without you in the room. And, as the delegation literature insists, hand over authority with the responsibility, not just the task, so a question can be resolved without routing through you. The test of whether it worked is simple and a little uncomfortable: take a real week off, fully unreachable, and see what breaks. What breaks is the map of what only lived in your head.
The reframe that makes this stick is that answering at midnight was never the generous act it felt like. The generous act is building a team that does not need the midnight answer, and the only way to build it is to stop supplying it. The founder who is always available is, in the most literal sense, training the people around them to be less capable than they are. The quiet of an inbox that the team has learned to handle without you is not neglect. It is the thing you were supposed to be building the whole time.