Almost nothing about this decision is urgent, which is unusual and is worth stating first: the cost of waiting six months is close to zero and the cost of starting badly is a year.
Four conditions worth having before you start
The process is stable. If the way the work is done is changing weekly, any documentation written this month is wrong next month and the person you hire is learning something that keeps moving.
Somebody has the time. A day a week for the first month, from a person who understands the work. If that person is in the middle of something else, wait for them rather than substituting somebody who cannot answer the questions.
The volume is real. Measured rather than estimated, by the method in the entry on volume.
It is not your peak. Starting an onboarding during the busiest quarter guarantees that the briefing does not happen and the person spends six weeks underused.
Lead times, so the calendar is realistic
Recruitment against a written role typically runs several weeks rather than several days, and longer for specialised skills. Candidates in employment serve notice, which in this country is commonly a month. Add your own interview scheduling.
From a signed engagement to a person at a desk, plan on a couple of months rather than a couple of weeks. A provider promising materially less is either sitting on a bench of unplaced staff, which raises its own questions, or is optimistic.
When to wait deliberately
When a system migration is underway or imminent. When the person who would brief is leaving or on leave. When the volume is short and the growth case is unproven. And when the actual problem is that a process is broken, because moving a broken process makes it harder to fix rather than easier.
The pressure to start immediately
It usually comes from a budget cycle or from a departure. Neither is a good reason and the second is a common one: somebody resigns, the role is expensive to replace locally, and offshoring becomes the answer to a gap rather than a considered decision.
That can work. It works better when the departing person writes the handover before they go, which is the one thing a notice period is genuinely good for and which almost nobody uses it for.
Sequencing more than one role
One at a time, with the second starting after the first is running without daily intervention. Hiring three simultaneously triples the briefing load in the same month, and the briefing load is the binding constraint rather than the cost.
Businesses that hire three at once usually conclude that the third was a poor hire. Frequently all three were fine and the attention was divided by three.
The review date
Set one before you start, ninety days out, with criteria written in advance: what the person should be doing without supervision by then, what your own time commitment should have fallen to, and what would constitute the arrangement not working.
Written in advance, that review is an honest assessment. Written afterwards, it is a justification, and the difference is visible to everybody including the person being reviewed.
The one thing that makes waiting expensive
Nothing, in most cases. The exception is where the work is currently being done by somebody who is about to leave and is the only person who knows it, in which case the clock is real and the priority is the handover document rather than the hire.
Starting in a quiet quarter
The best time is a period when the work exists steadily and nobody is under pressure. For many businesses that is a specific and predictable stretch of the year, and planning around it costs nothing.
The second best time is now, if the four conditions hold. There is no prize for waiting once they do.
The pilot question
Clients frequently ask for a trial period, and providers frequently offer one. It is reasonable and it measures less than people expect: a three-month trial ends roughly when the person becomes productive, so it tests the onboarding rather than the arrangement.
If you want a trial, make it six months and write down in advance what it is testing. A trial without stated criteria becomes a conversation about impressions.
Notice on your side
Check what leaving costs before you start, not because you expect to leave but because the answer tells you how confident the provider is. Symmetric notice is a reasonable ask and an unsymmetrical arrangement is worth a question.
The handover a departure buys you
A notice period is the one moment when somebody who knows a process has both the knowledge and the time to write it down. Almost every business spends it on recruitment instead.
Ask for the document in week one of the notice rather than week four. By week four the person has mentally left and the document shows it.
Announcing it internally
Before rather than after. A team that discovers the arrangement by meeting a new colleague on a call assumes the worst about what it means for them, and that assumption is expensive to correct.