Mental load: what founders carry in their heads, and why it never ends

Mental load is everything a founder or executive carries in their head that no one else carries for them: anticipating what will be needed, choosing, deciding and checking that it happened. It is not measured in hours, so it is easy to miss. It still sits on you all day, follows you home and takes focus from the work that matters most.

What is mental load?

Mental load is the cognitive work around execution. Not booking the flight, but remembering it needs booking, comparing options, deciding and confirming the ticket actually arrived.

Harvard sociologist Allison Daminger broke this work into four stages (American Sociological Review, 2019):

1. Anticipating a need before it becomes a problem.
2. Identifying the options.
3. Deciding.
4. Monitoring until it is done.

Her research looked at households. For founders and business owners the mechanism is the same, with a longer list: vendors, contracts, insurance, travel, authorities, collections, the home and the family. Each item is small. Together they form a layer nobody sees.

A normal week for a business owner can look like this. On Sunday the accountant writes about a missing document. On Monday a vendor needs approval on a quote. On Tuesday you remember one of the kids' passports has expired before the holiday trip. On Wednesday a client still has not paid, and you have to decide whether to call yourself. None of these takes more than a few minutes. But each one opens in your head the day it arrives and stays open until someone closes it. By the end of the week the list is longer than it was at the start, even after sixty hours of work.

Why doesn't it go away at the end of the day?

Because the brain does not close open loops. It keeps reminding you of them.

A 2011 study by E.J. Masicampo and Roy Baumeister showed that an unfulfilled goal keeps intruding on your thoughts and hurts performance on unrelated tasks. The striking finding: making a concrete plan for the goal was enough to stop the interference, even before the task itself was done (Journal of Personality and Social Psychology, 2011).

Sophie Leroy of the University of Washington named a related effect "attention residue". When you move on from a task that is not finished, part of your attention stays behind, and performance on the next task drops (Organizational Behavior and Human Decision Processes, 2009).

That is what happens in the middle of a board meeting when you remember the car insurance expires tomorrow. Or at ten at night, when an unsigned contract will not let the day end. The problem is not the time the task takes. It is the attention it takes before anyone has started on it.

Why are founders sure it is faster to do it themselves?

Almost every founder I know says the same sentence: "By the time I explain it, I could have done it myself." For a single task, they are right.

The trouble starts the second time. You explain it to someone, and next time they have forgotten. Or they leave, and you explain everything again to someone new. Each round teaches you that delegating costs more than it saves. So you take it all back, and the list in your head gets longer.

The data says this habit is expensive. Gallup studied 143 CEOs from the Inc. 500. Those with strong delegation talent grew 1,751% over three years, 112 percentage points more than those who struggle to delegate, and generated 33% more revenue (Gallup, 2015).

So the problem is not that leaders refuse to delegate. It is that delegation does not stick.

Why don't AI agents solve this?

The last two years added another promise: you no longer need to delegate to anyone, just ask the agent. Many leaders I talk to are sure the problem is already solved. For most of them, the opposite happened.

Back to Daminger's four stages. An AI agent is very good at the second one, identifying options. It can compare prices, draft an email and fill in a form. The other three stay with you:

• Anticipating. The agent does not know your insurance renews in two weeks. Someone has to remember to ask it.
• Deciding. It brings three options and hands the decision back to you, sometimes with new questions.
• Monitoring. It sent the email. Did the vendor reply? Was the form accepted? Someone has to check.

So one task becomes several sub-tasks. First explain to the agent what you need, with all the context. Then check what it did. Then handle what it could not do. Each sub-task is another open loop in your head, exactly the kind Masicampo and Leroy describe. The agent saved execution time, and the mental load stayed with you.

There is another problem. Next time, everything starts from zero. Whether you handled it yourself or asked an agent, when the task comes back next year it comes back to you. You have to remember it exists, explain it again and follow up again. No one but you carries it over time.

That is the difference between a tool and an owner. A tool does what you asked. An owner remembers to ask, decides what can be decided without you, makes sure it happened, and handles it again next time. We wrote more about this difference in Ankora vs AI assistants.

What did 25 years of leading companies teach me?

I have led technology companies for more than 25 years. I founded Seemore and Aggua and served as CEO of both, and held senior leadership roles at Verbit and Feedvisor. In every one of those roles I carried the same list in my head. It was never part of my job, and it was always there.

I tried almost everything. Task apps that helped me remember but took nothing off my plate. Separate service providers, each solving one piece and leaving me to coordinate between them. Good people who took all the context with them whenever they moved on. And lately AI tools, which saved execution time but left everything else with me. Every time I reached the same conclusion: what was missing was not another pair of hands. It was someone who holds the list instead of me, and remembers.

How do you remove the load without losing control?

We built Ankora around the two things I was missing: one accountable person, and memory that does not get lost.

A dedicated human Operations Manager. One person who owns the outcome end to end. They decide what is right, what is urgent and what truly needs you. Financial, sensitive or unusual decisions always go through a person.

AI orchestration with persistent memory. Our technology layer keeps every preference, every decision and every vendor you have worked with. You explain once. Even if your Operations Manager changes, the context stays. The system also tracks every open process and flags a delay before it becomes a problem.

In practice it is simple. You send a request however suits you: a call, a message or an email. The Operations Manager decides what is right and what needs you, the AI layer holds the context and tracks it, and a network of vendors and service providers executes under supervision. The next request is shorter than the last, because the system already knows.

Here the AI works for you, not the other way around. You are not the one explaining to it and checking its work. The Operations Manager is, and you receive only what truly needs you. When a task comes back, it is already known, and handled before you remember it.

This is the link the research points to. Every task gets a plan and an owner the moment it arrives, so your mind can let go of it. Delegation lives in the system's memory, not in one person's.

What can you do this week?

1. Write down the list in your head. Everything open, even the small things. Writing it down already relieves some of the pressure.
2. Mark what truly needs you. It is usually less than a fifth of the list.
3. Give every other item an owner and a date. A task without an owner comes back to you.
4. Write context once, in one place. Preferences, vendors, recurring details. The next explanation gets shorter.
5. After two weeks, check what came back. Whatever returned shows where delegation did not stick.

If the list is too long to manage alone, that is exactly what Personal Operations Management for founders is built for.

The goal is not to do more. It is to think about less.