What to hand off to an AI assistant first

Why the order matters

Most owners who try an AI assistant start in one of two ways. Some hand it a single small task, like rewording an email, and decide it isn't worth the bother. Others try to hand it everything at once and stop the first time it gets a client's name wrong. Both miss the useful middle: give it a clear set of jobs that come up every week, check its work closely for a while, and widen what it handles as it proves itself. That's how you'd bring on a new hire, and it works for the same reasons.

Three questions sort a job into "hand it off now", "hand it off later" or "keep it":

• Does it repeat? Work that comes up every week is worth setting up once.
• Can it be drafted? If the output is a message, a document or a summary, an assistant can prepare it.
• Is it quick to check? If you can tell in under a minute whether the draft is right, the risk of handing it off is low.

A job that scores yes on all three is a good first hand-off. A job that commits money, a date or a relationship stays with you, however often it repeats.

Week one: the jobs to hand off first

For a small service business, such as a planner, a venue, a florist or a booking agency, these are the usual starting points:

1. Replies to new inquiries. The first answer to a new lead is mostly the same facts in a slightly different order: availability, what you offer, the next step. A drafted reply in your voice, waiting for your yes, removes the delay without removing you.
2. Follow-ups on open quotes. Proposals often go quiet because nobody sends the second follow-up, not because the client said no. Your assistant can keep the list of open quotes and draft each follow-up when it's due. Our guide to the follow-up email after a quote has the day 3, 7 and 14 pattern.
3. Meeting write-ups. After a client call, a write-up of what was agreed and what happens next takes you minutes to check and saves the "what did we say about the timeline?" email two weeks later.
4. A morning brief. One message at the start of the day: what's waiting on you, who needs a reply and what's on the calendar. It costs nothing to check and shows you quickly how well the assistant understands your business.
5. Keeping track of what's open. Unanswered inquiries, unsigned contracts, unpaid deposits. The list itself is the job; the assistant keeps it current and drafts the next nudge.

Each of these leaves you as the person who decides what goes out. With Gina, every draft waits in one inbox of approvals until you approve, edit or skip it.

What to keep

Some work should stay with you or your team however good the drafts get. An AI assistant can prepare it, but the decision is yours:

• Pricing and exceptions. What to charge, when to discount and when to walk away.
• Commitments. Holding or releasing a date, signing a contract, sending a payment request. Gina can draft contracts and payment requests; none should go out without your yes.
• Hard conversations. An unhappy client, a vendor who let you down, a change of plan on the day. The assistant can pull together the history so you walk in prepared; the conversation is yours.
• Anything you can't check quickly. If verifying a draft would take as long as writing it, it isn't a good hand-off yet.

How to widen it over time

Trust in an assistant should come from its record, not its promises. A simple way to build it:

1. Approve everything for the first few weeks. Read each draft. Edit what's off and tell it to remember the fix.
2. Write down what you keep correcting. Tell your assistant to remember it: your sign-off, the prices you quote, the clients who prefer a phone call. Gina keeps those in one memory your team shares.
3. Move routine work onto a schedule. Once a job is reliable, set it to run on its own: the weekly check of open quotes, the morning brief, the write-up after each meeting. The results still come to you.
4. Pre-approve only what has earned it. If a reply type has been right every time for a while, you can choose to let it go out without a check. Keep money, dates and contracts on approval.

What not to expect

An AI assistant only sees the apps you connect and can't do more than the access you give it. It doesn't replace your CRM or your accounting tool; those stay the source of truth. And it can be wrong, which is exactly why the approval step exists. Used this way, the assistant does the preparing and the tracking, and you keep the decisions.

If the idea is new, start with What is an AI chief of staff?, and before you connect your inbox, read Is it safe to give an AI assistant your email and calendar?