Day 1: the founder hands over a task list. Answer emails, manage the calendar, help with scheduling. Three lines, ten minutes, nobody writes down what good looks like beyond that, and that gap is usually why virtual assistants quit by day 90.
Day 30: the founder is frustrated. “They’re not getting it.” Small things are being missed, decisions are being made that the founder wouldn’t have made, and it feels like managing the assistant is costing more time than doing the work would have.
Day 60: the assistant is frustrated too, just as quietly. “They keep changing their mind.” Priorities that were clear a month ago have shifted without explanation. Instructions that felt final turn out to be provisional. What looked like a straightforward brief now looks like guesswork.
Day 90: someone leaves. Usually the assistant, sometimes the founder ends it first. Both walk away with a story about what went wrong, and the two stories don’t overlap. Neither side is lying. Neither side wrote down what success looked like, and that’s the part both stories miss.
Why Virtual Assistants Quit: What Day 1 Actually Needed
The task list is never the problem by itself. Answer emails, manage the calendar, that’s a fine starting brief, and the tasks most founders hand over first are usually the right ones to start with. When an investor email lands at 11pm, does it wait until morning or get flagged now. What does “urgent” mean to this specific person, as opposed to what merely looks urgent on the page.
None of that shows up in a task list, because it lives in the founder’s head as instinct, not instruction. A fifteen-minute conversation at the start, walking through five or six real decisions the assistant is likely to face and how the founder would call them, does more for the first month than any task list could. It rarely happens, because writing down your own instincts feels like exactly the kind of thinking you hired someone to take off your plate.
What “they’re not getting it” is actually describing
By day 30, “not getting it” nearly always means one specific thing: the assistant made a call the founder wouldn’t have made, and the founder is now doing double the work, one to fix the decision, one to explain what should have happened instead. From the founder’s side this reads as a competence problem.
From the assistant’s side it usually isn’t. It’s an information gap. They made the most reasonable call available with what they knew, and what they knew was incomplete because nobody had walked them through the founder’s actual priorities yet. The fix at day 30 isn’t more oversight. It’s the conversation that should have happened at day 1, finally happening, thirty days later than it needed to.
What “they keep changing their mind” is actually describing
By day 60, the assistant has usually stopped asking questions. Early on, asking felt normal. By now it feels like proof they haven’t figured out the job, so they guess instead, and guessing under pressure to look competent produces worse guesses than asking would have. “They keep changing their mind” is what that looks like from the outside: work that used to align with expectations now consistently misses, because the assistant switched from checking to assuming.
Meanwhile the founder has usually started double-checking work that should be trusted by now, which the assistant reads as a lack of confidence in them, which pushes them further from asking. Both sides are reacting to the same silence and drawing different conclusions from it.
Why day 90 doesn’t have to be the ending
Skip it, and ninety days in, you’re not looking at a bad hire. The hiring decision may well have been the right one. You’re looking at two people who agreed on almost nothing about how the job actually works.
Onboarding an assistant isn’t an HR step to complete once and file away. It’s the mechanism that turns a stranger doing tasks into someone trusted with judgment, and it has to be revisited, not just written once at the start. Skip it, and ninety days in, you’re not looking at a bad hire. You’re looking at two people who agreed on almost nothing about how the job actually works, and never found out until it was too late to fix.
If you’re about to bring someone on, the question worth answering before day 1 isn’t what tasks you’ll hand off. It’s whether you could sit down and describe, out loud, the last five decisions you made for your business and why you made them that way. If you can’t do that yet for yourself, that’s the actual first task, before any task list.
Common questions
Why do virtual assistants quit around 90 days?
Because neither side wrote down what success looks like. By day 30 the founder sees missed calls, by day 60 the assistant sees shifting priorities, and both stories describe the same gap from opposite ends.
What should a founder give a new assistant on day one?
More than a task list. Spend fifteen minutes walking through five or six real decisions the assistant will face and how you would make each one, such as whether an investor email at 11pm waits until morning or gets flagged now.
What does “they’re not getting it” usually mean at 30 days?
The assistant made a reasonable call with incomplete information. It’s an information gap, not a competence problem, and the fix is the day-one conversation that never happened, not more oversight.
Why do assistants stop asking questions?
By day 60, asking feels like proof they haven’t learned the job, so they guess instead. Guesses made under pressure miss more often, and from the outside it looks like the founder keeps changing their mind.
Can a struggling assistant relationship be saved before day 90?
Yes. The hire is often the right one. Go back to onboarding: write down how you make decisions and walk through it together. Onboarding has to be revisited, not completed once and filed away.
What should a founder do before hiring an assistant?
Check whether you can describe, out loud, the last five decisions you made for your business and why you made them. If you can’t yet, do that first, before you write any task list.