C-level Executive

Smaller than expected at first, and awkward for about a month. A founder who built the firm has every process in their own head. The first weeks are spent getting things written down that never were. Practically, it starts with the calendar and the inbox, because those are visible and correctable. Then travel, expenses, and meeting preparation. Then follow-up on commitments, which is where most founders quietly lose things. Only after that does the work become anticipatory. Expect to spend more time than you save in the first two weeks. Founders who abandon the arrangement almost always abandon it in that window. It is the wrong moment to judge it, and it is when almost everybody does. Set the review point at the end of the second month, and hold to it through the first weeks.

The logistics and the preparation, not the content. That covers scheduling across several senior calendars, which is genuinely difficult work. It covers assembling the board pack from material other people owe you, and chasing those people. Then distributing documents, taking minutes, and tracking action items to closure between meetings. Reporting packs, KPI dashboards, and the recurring partner updates are prepared from the numbers your finance function produces. Nothing is assembled from scratch each quarter. What stays with you is what the material says and what is committed in the room. Handling this class of information is exactly why confidentiality is part of the job description rather than an add-on. Access to board and investor material follows the same least-privilege rule as everything else. Set it up before the first pack rather than after it.

Through the contract, the access controls, and the selection, in that order. Confidentiality obligations are written into the engagement. Handling sensitive communication is part of the job description for this role rather than an exception to it. Access follows least privilege: the assistant sees what the work requires and no more, through named accounts in your own systems. Some material simply should not be delegated, and deciding which is your call, not ours. Practically, the categories worth thinking about in advance are compensation, legal matters, anything under a client NDA, and personnel issues. Firms that decide those boundaries at the start have a much easier time of it. The alternative is deciding them in the moment. That means the first time something uncomfortable lands in the inbox, on a Friday afternoon.

Anything that is the relationship itself. Client conversations that carry the firm's reputation. The calls where you set direction and the decisions your name is on. And any communication where the recipient needs to hear it from you. Also anything you cannot yet articulate, because a task you cannot describe is a task that will come back wrong and cost you twice. Delegation works on things with a definable outcome. It fails on judgment you have not yet made explicit, and it fails quietly rather than obviously. The useful test is simple. Would you be comfortable telling the recipient who actually did it? If the answer is no, keep it. Everything else moves fine once the context has been transferred. That includes a surprising amount of what feels irreplaceable at the start of an engagement.

Longer than a week and shorter than most people fear, and it depends almost entirely on how much context you transfer. Once you have selected the person, onboarding for a standard profile takes less than 72 hours. That is when the work starts, not when the benefit arrives. The pattern is consistent. The first two weeks cost time and the second two are roughly neutral. The return comes in the third, when the assistant starts anticipating instead of responding. What accelerates it is explaining the reasoning behind corrections and letting them see how decisions get made. What slows it down is only ever handing over fully specified tasks. Ask us what the first month actually looks like before you commit to a schedule. Judge the arrangement at week six rather than week one.