ChatGPT for HR: Recruitment, Onboarding, and Policies

HR teams are drowning in documentation. Job descriptions, interview guides, onboarding checklists, policy documents, performance review templates — most of this writing follows well-established formats that take significant time to produce from scratch every time. ChatGPT doesn't replace HR judgment, but it compresses the document production work dramatically.

I've spent time with HR teams at companies from 15 to 5,000 employees, and the pattern is consistent: the work that gets deprioritized is always the writing. The offer letter that went out without the updated benefits language. The job description copy-pasted from three years ago. The onboarding checklist that hasn't been touched since the last HR manager left. ChatGPT helps address this.

Here's how to use it across the main areas of HR work.

Job Description Writing

A well-written job description does two things: attracts candidates who are a genuine fit and screens out those who aren't. Generic JDs written from templates tend to attract too broad a pool or signal nothing meaningful about the role.

Full job description prompt

"Write a job description for a [job title] at a [company stage: early-stage startup / mid-size company / enterprise] in the [industry] industry. The role is [remote/hybrid/in-office]. Responsibilities: [list 5-8]. Must-have requirements: [list]. Nice-to-have: [list]. Salary range: [range]. Include: a company description paragraph that doesn't sound generic, a 'what success looks like in 90 days' section, and a note on our culture that speaks to [specific cultural values]. Avoid corporate filler phrases."

Bias audit prompt

"Review this job description for: gendered language (e.g., 'rockstar', 'ninja', 'aggressive'), unnecessary degree requirements, requirements that could screen out qualified non-traditional candidates, and cultural fit language that's vague or exclusionary. Suggest specific replacements."

Common findings: "must be able to work in a fast-paced environment" (vague and meaningless), "bachelor's degree required" (often unnecessary for the actual role), "strong communication skills" listed as a requirement (every JD says this; it adds nothing).

Role leveling prompt

"I need to differentiate a job description for three levels of the same role: [role title] I, II, and III. The main differences are: [describe scope, autonomy, impact expectations]. Write three separate 'responsibilities and qualifications' sections that clearly distinguish the expectations at each level."

Interview Question Generator

Structured interviews — where every candidate answers the same questions — produce better hiring decisions. Unstructured interviews favor candidates who interview well over candidates who perform well. ChatGPT is good at generating structured question sets.

Behavioral interview questions

"Write 8 behavioral interview questions for a [role] position. Focus on these competencies: [list 3-4: e.g., ownership, communication, problem-solving under pressure, collaboration]. For each question, include: the question itself, what competency it tests, and 2-3 points that indicate a strong answer. Use the 'Tell me about a time...' format."

Technical screening questions

"Write 6 technical screening questions for a [role: software engineer / data analyst / marketing manager / finance analyst] candidate with [experience level] experience. Include a mix of: knowledge verification questions, scenario-based questions, and one question that has no single right answer to test thinking process."

Culture and values questions

"Write 5 interview questions that assess alignment with these company values: [list values]. Each question should reveal actual behavior and decision-making, not prompt candidates to recite values back at us."

Reference check questions

"Write 10 reference check questions for a [role] hire. Include questions that surface: work quality, collaboration style, how the person handles criticism, what they're genuinely best at, and where they've struggled. Make them conversational, not formal."

One thing worth noting: interview questions related to protected characteristics (age, national origin, pregnancy, religion, etc.) can create legal liability. The SHRM resource center has guidance on legally compliant interviewing if you're unsure about specific questions.

Onboarding Checklist and Plan

New hire onboarding failure is a real problem — studies consistently show that poor onboarding increases early turnover. A structured 30-60-90 day plan helps new employees feel oriented and actually retain more of what they learn.

Onboarding checklist prompt

"Create a comprehensive onboarding checklist for a new [role] at a [company type]. Organize by timeline: Pre-Day 1, Day 1, Week 1, Month 1, Month 2, Month 3. For each period, include: systems access and setup tasks, key meetings and introductions, documents to review, goals to achieve, and who is responsible for each item (HR, manager, buddy, or the new hire themselves)."

30-60-90 day plan prompt

"Write a 30-60-90 day onboarding plan for a new [role]. Day 1-30 focus: learning and observing. Day 31-60 focus: beginning to contribute. Day 61-90 focus: working independently. For each phase, include: 3 learning goals, 2 relationship goals, and 1 deliverable that marks successful completion of that phase."

Manager onboarding guide

"Write a guide for managers on how to onboard a new team member effectively. Cover: what to do before they start, how to structure the first week, how to give early feedback, common mistakes managers make in onboarding, and how to set up a weekly check-in rhythm. Tone: practical, not corporate."

For broader employee experience considerations, the ChatGPT prompt bible covers communication prompts that work well for internal HR announcements and culture-building messages.

Policy Drafting

HR policies are one of the most valuable use cases for ChatGPT in HR — the structure is well-defined, the language is formal, and the first draft is the hardest part. Always have legal review the final version.

Policy draft prompt

"Draft an HR policy on [topic: remote work / time off / expense reimbursement / performance improvement / workplace conduct]. Include: purpose statement, scope (who it applies to), policy details (the actual rules), employee responsibilities, manager responsibilities, and how violations are handled. Tone: clear and direct, not legalistic. Flag any sections where local employment law variation may apply."

Employee handbook section prompt

"Write the [welcome / benefits overview / code of conduct / communication guidelines] section for an employee handbook. Company size: [size]. Industry: [industry]. Tone: [warm and direct / professional and formal]. Keep each section under 400 words. Avoid corporate jargon."

Policy gap finder

"Here are the HR policies our company currently has: [list]. What standard HR policies are we missing for a [stage/size] company in the [industry] industry? Flag any gaps that represent significant compliance or people risk."

Performance Review Templates

Performance reviews are one of HR's most time-consuming documents — and one of the most consistently complained about by both managers and employees. A good template gives structure without requiring managers to write from scratch.

Review template generation

"Create a performance review template for [role category: individual contributor / manager / executive]. Include sections for: overall rating with definitions (e.g., what does 'meets expectations' actually mean?), goal achievement review (with space for prior goals and outcomes), key strengths with behavioral examples, development areas with specific improvement suggestions, and goals for the next review period. Include guidance notes for managers on how to complete each section."

Review calibration guide

"Write guidelines for managers on how to write effective performance feedback. Cover: the difference between vague and specific feedback, how to document performance with examples, how to give critical feedback constructively, legal considerations in written reviews, and common mistakes that can undermine a review's effectiveness."

Promotion justification prompt

"Help me write a promotion justification memo for [name/role] being considered for [target role]. They have demonstrated: [list achievements and behaviors]. Comparable market data suggests [salary/title adjustment]. Write a structured memo that makes the business case clearly and connects performance to organizational impact."

Compensation and Offer Letter Templates

Salary benchmarking prompt

"What data sources should I consult for compensation benchmarking for a [role] in [city/region]? List both free and paid options with what each covers. Also describe the methodology for setting a pay range using percentile positioning."

Offer letter template

"Write an offer letter template for a [role] position. Include: job title and reporting relationship, start date placeholder, base salary, equity (placeholder for terms), benefits summary line, at-will employment statement (for US), contingencies (background check/reference check), and a warm closing. Tone: professional but welcoming."

HR Metrics and Reporting

HR dashboard prompts

"What are the most important HR metrics for a [size] company to track? Organize by category: recruiting, retention, engagement, and DEI. For each metric, include: definition, how to calculate it, and what benchmark I should target."

"Write an HR quarterly report template. Include sections for: headcount changes, recruiting metrics (time to fill, offer acceptance rate), retention and turnover analysis, engagement highlights, and a priorities section for next quarter. Format: suitable for an executive audience."

For connecting HR workflows to broader business analysis, the ChatGPT code interpreter guide covers data analysis techniques useful for HR metrics and workforce planning.

What ChatGPT Won't Do Well in HR

A few cautions from practical experience:

Performance ratings: ChatGPT cannot determine whether an employee deserves a rating of 3 vs 4. That requires manager judgment based on actual observed work over time. Use it for language and structure, not assessment.

Difficult conversations: Scripts for firing, demotion, or delivering hard feedback generated by ChatGPT need heavy customization. These conversations depend on relationship context, individual circumstances, and delivery that no template captures.

Legal compliance: Employment law is jurisdictional, industry-specific, and changes frequently. ChatGPT's knowledge has a training cutoff and cannot give you current, jurisdiction-specific legal guidance. Always verify with a qualified employment attorney.

The ChatGPT vs Claude comparison is useful context if you're evaluating which AI model handles sensitive HR writing tasks — like difficult termination language or harassment policy drafts — with the right tone.

Conclusion

ChatGPT doesn't replace HR expertise. It replaces the blank page. Job descriptions, interview guides, onboarding checklists, policy documents, performance templates — these are documents with known structures that take hours to produce from scratch every time. With the right prompts, you get a solid first draft in minutes, leaving time for the judgment and relationship work that actually defines good HR.

Start with the job description prompt for a role you're currently hiring. Run it, compare it to your current JD, and see what's better and what needs adjustment. That fifteen-minute experiment will show you the value more clearly than any explanation.

Keep the legal review step non-negotiable. Everything else is a starting point.

Further Reading

• 10 ChatGPT Prompts That Write Better Emails Than You
• 15 ChatGPT Prompts for Cold Emails That Actually Get Replies
• How to Use ChatGPT Code Interpreter for Data Visualization
• ChatGPT for Game Design: NPC Dialogue, Quests, and Lore
• I Spent 100 Hours with ChatGPT-4o — Here's Everything I Learned