The First 90 Days: Onboarding Into a Small Business

Most small-business onboarding is a laptop, a set of logins, and a hope that proximity will do the rest. Then at month three everyone is quietly disappointed and nobody can say exactly when it went wrong.

The first ninety days is where a hire either works or doesn't, and it is almost entirely within your control.

Write the plan before the offer

If you can't describe what good looks like at day 30, the role isn't defined well enough to hire into — and the person you're about to make an offer to will spend a month guessing.

Three outcomes, one per month:

Day 30 — Running the weekly reconciliation independently, with queries raised rather than guessed.

Day 60 — Owning month-end close start to finish, with me reviewing rather than doing.

Day 90 — Close completed within three working days without my involvement, and one improvement to the process identified and implemented.

Send this with the offer. It tells a good candidate that you've thought about it, and it gives both of you something to measure against that isn't a feeling.

Before day one

• Everything works on the first morning. Email, systems, access, payroll set up. Someone spending day one waiting for logins has learned something about how the business runs.
• The 30, 60 and 90 day conversations are in the calendar. Put them there now. The ones that get scheduled later are the ones that get postponed, and the ones that get postponed are the ones where something is wrong.
• A written first week, hour by hour for day one and day-level after that.

Week one — ship something

The single highest-value onboarding decision: week one ends with something real, however small, that the new person did and that is now live.

A reconciled account. A published page. Three customer emails answered. It doesn't matter what it is, only that it's genuine and it's finished.

Nothing else builds confidence like having already contributed, and nothing makes the first week feel more pointless than five days of reading and shadowing.

Around that, week one covers:

• Context first, tasks second. Who the customers are, what they buy, why it matters. A person who understands the business makes far better small decisions than one who has only been shown which buttons to press.
• One named go-to for questions. In a five-person business that's usually you. Say it explicitly, and say that early questions are expected rather than a sign of struggling.
• The working norms, stated. How feedback happens, what gets written down, response expectations, when you're not available.

Weeks two to four — document as they learn

Here's the move that converts a weakness into an asset.

Have the new person write the SOP for each process as you teach it. You get documentation you didn't have, they learn actively rather than passively, and you discover very quickly how clearly you actually explain things.

Use the SOP format — trigger, owner, one action per step, explicit definition of done. Their draft is testable immediately: hand it back and have them run the process from their own document.

By week four you should be delegating properly — outcome, constraints, decision level, checkpoint — rather than assigning tasks.

Expect a real ramp

| Month | Realistic expectation | |---|---| | 1 | Limited independent output. Consumes meaningful time from you. | | 2 | Partial contribution. Still checking in on non-routine decisions. | | 3 | Approaching full contribution on the defined outcomes. |

Specialist or client-facing roles run longer. Plan your cash for that, and add the hire to the 13-week forecast at full cost from day one — because payroll is fully loaded from day one whatever the output is.

If your plan only works when the hire is productive in week two, the plan doesn't work.

The 30, 60 and 90 day conversations

Thirty minutes each, and the same four questions every time:

1. Against the outcome for this month, where are we?
2. What's getting in your way?
3. What has surprised you — good or bad?
4. What do you need from me that you're not getting?

Question four is the important one, and you have to actually act on the answer or people stop giving you real ones.

Be specific. "Going well" is not feedback. If something is off, name it at 30 days rather than hoping — most 30-day problems are onboarding failures, and onboarding failures are recoverable when they're named early.

The mistakes

1. No written plan. Both of you spend a month guessing.
2. A week of shadowing. Passive, and it teaches nothing that sticks.
3. No named go-to. They either interrupt everyone or guess quietly.
4. Postponing the 30-day conversation. Precisely when it matters most.
5. Expecting productivity in week two. Then reading a normal ramp as a bad hire.
6. Teaching without documenting. You'll teach it again to the next person.

What to do next

Write the three monthly outcomes today, even if the hire is months away. If they don't come easily, that's your finding — the role isn't defined, and defining it is cheaper now than after someone has started.