A thirty, sixty and ninety day plan that is not a formality
The value of the plan is not the document. It is that both parties have agreed in advance what progress looks like, before either can be disappointed by surprise.
Ninety-day plans have a poor reputation because most of them are written to be filed. A page of aspirations, produced in the first week, referred to never, and remembered only at the probation meeting when someone goes looking for evidence.
Used properly the plan does something specific and valuable: it converts a vague mutual expectation into a written one, early enough that a mismatch can be corrected rather than discovered.
Three phases with different purposes
The distinction between the phases is not how much work is expected but what kind.
The first thirty days are about learning and context: understanding the work, the customers, the systems, who does what. The output is comprehension, and the plan should say so rather than pretending otherwise. Expecting substantial independent delivery in month one is how small employers acquire a reputation for throwing people in. For broader independent guidance on this point, consult OPM performance-management resources.
Days thirty to sixty are supervised delivery. The person is doing the actual job with a safety net — work reviewed before it goes out, decisions checked, a defined person to escalate to.
Days sixty to ninety are independent operation on the core of the role, with the manager's involvement dropping to exception handling. If that transition has not happened by ninety days, something specific is wrong and the plan makes it visible.
'Attend three client calls' is an activity and will be completed regardless of whether anything was learned. 'Able to run a routine client call without support' is an outcome, and it can be assessed.
Write it with the person, in week one
A plan handed to someone is a set of instructions. A plan written together in the first week is an agreement, and the difference shows in whether it is ever mentioned again.
It also surfaces mismatches immediately. If the new hire expects to be leading a workstream by month two and the plan describes supervised delivery, that conversation is far better held in week one than in month three.
Keep it short and specific
Three to five outcomes per phase. More than that and it stops being a plan and becomes a description of the job.
Each outcome needs to be assessable by someone other than the author. 'Understands our pricing' is not; 'can quote a standard job without checking' is. The test is whether the manager and the employee would independently agree on whether it has been achieved.
Review it on the dates, not at the end
The reviews are the mechanism. Thirty minutes at day thirty, day sixty and day ninety, in the calendar from week one so they are not arranged under pressure.
Each review runs the same way: what has been achieved, what has not, what is blocking it, what changes for the next phase. Most of what surfaces is fixable and small — an access request nobody actioned, a handover that never happened, a piece of context everyone assumed the person had.
It runs in both directions
The plan should record what the employer committed to as well: the training that would be provided, the access that would be arranged, the time the manager would make available. For a related product-level perspective on time management, see here.
Small employers underestimate how often a new hire's slow start is caused by an unmet commitment from their side. A plan that only lists the employee's obligations produces a review conversation in which the employee is accountable for the consequences of somebody else's delay.
Use it to end things honestly, if it comes to that
Occasionally an appointment is not going to work. A documented plan with recorded reviews means that conclusion is reached earlier, on evidence, with the person having had a fair opportunity to address specific gaps.
That is better for everyone than the alternative, which is a manager privately deciding at week six and saying nothing until the probation meeting. It is also, in most jurisdictions, the difference between a defensible decision and one that is not.