Hiring & People · Updated July 17, 2026
ChatGPT prompt for a 30-60-90 onboarding plan
This is the prompt we run once the offer is accepted, to build the onboarding plan before day one. It turns the scorecard into a one-page 30-60-90 plan. Its rule: it derives milestones only from your scorecard outcomes, and flags any outcome that can’t realistically be assessed by day 90.
The prompt to copy
Create a 30-60-90 day plan for our new {role}, starting {date}. Their scorecard outcomes: {paste}. Team: {names/roles}. Tools: {list}. Key processes they must learn: {list or links}.
Format per phase: goals, what they'll actually do, who they'll meet and why, what "on track" looks like (measurable where possible — derive only from the scorecard outcomes I gave), check-in schedule with me.
Flag any scorecard outcome that can't realistically be assessed by day 90, so I set expectations instead of guessing. One page.
Replace the role, the date, and the fields in curly braces, paste the scorecard from H1, then run it: the plan comes out on one page, ready to share with the hire.
When to use it
Once the offer is accepted, before day one. A plan ready on day one turns the first weeks into a clear trajectory, instead of a drift where the hire waits to be told what to do.
What you get back
- Three phases, 30 then 60 then 90 days, with goals.
- The concrete: what they’ll do and who they’ll meet.
- “On track” milestones, measurable and drawn from the scorecard.
- The outcomes that can’t be judged by day 90, flagged in advance.
Real example: a plan honest about its horizon
Input: Maya, at the Brightside agency, pastes the project-lead scorecard, starting {date}, with the team (Tom, Lena, Priya) and the tools. Output excerpt:
90-day phase — goal: autonomy on routine client projects.
On track: 2 projects run solo, delivered on time.
To flag: the outcome "take back the 3 key accounts" can't be judged by day 90 —
the client trust cycle runs longer. To confirm together as a 6-month goal,
not a 90-day one.
Notice the last block: the AI didn’t pretend everything is measurable at 90 days. It flagged the longer-horizon outcome so Maya sets expectations from the start. That is the prompt protecting the hire from an unfair review.
The full 30-60-90 plan, with all three phases detailed and the P7 playbook that chains it from the scorecard, ships inside the kit.
Frequently asked questions
- Does this prompt work with Claude or Gemini?
- Yes. It relies on no ChatGPT-specific feature: paste it as-is into Claude or Gemini and the result is the same.
- Why does the plan flag what can't be assessed by day 90?
- Because judging an outcome that takes six months to show is unfair to the new hire. The prompt names it so you set expectations from the start, instead of discovering the mismatch at the first review.
- Does the AI invent the "on track" milestones?
- No. The prompt derives them only from the scorecard outcomes you paste. It fabricates no performance goal you didn't already define at hiring time.