Spot Fall-Off Risk Between Accept and Start Date
When to use it
Work out which accepted candidate isn't going to show up, while there's still time to do something about it.
The prompt
{{Candidate Name}} accepted our offer for {{Job Title}} on {{date}} and starts on {{start date}}. Assess the risk that they don't start. HOW THE ACCEPT WENT: {{immediate, took a week, needed a bump}} NOTICE PERIOD: {{length, and whether they have actually resigned yet}} OTHER PROCESSES THEY WERE IN: {{what stage, and whether they've told those companies}} CONTACT SINCE ACCEPTING: {{how often, who initiated, how fast they reply}} WHAT THEY WERE OPTIMISING FOR: {{money, scope, manager, stability, location}} ANYTHING THAT'S CHANGED OUR SIDE: {{their manager left, the team reorganised, the start date moved}} Give me: 1. A risk read with the specific reasons behind it, not a score on its own. 2. Which signals in what I told you actually predict a fall-off, and which are just a quiet person being quiet. 3. A contact plan across the notice period: who reaches out, how often, and what each touch is for. 4. What their new manager should do, and when — this matters more than anything I do. 5. The one honest question worth asking directly, and how to ask it without planting the idea. 6. What to have ready if they do fall off: who's next, and whether that pipeline still exists. Don't tell me it's fine because they seemed enthusiastic. Enthusiasm at accept isn't predictive.
Tip
The manager's contact matters more than the recruiter's. Someone wavering under a counter-offer is choosing between two teams, and only one of them has actually spoken to them since they said yes.
How to use it
Paste the prompt into ChatGPT, Claude or whichever assistant you use, then replace every {{bracketed}} part with your own detail. The more specific and messier your input, the better the output — a model given raw notes has more to work with than one given a tidy summary you wrote first.
More Offers & Feedback prompts
Open the full tool for all 87 prompts, the glossary, and the JD decoder.