Build a Team That Scales With You
AI for Hiring and People Management
It's Wednesday afternoon. You've posted a job for a marketing coordinator, and 94 applications landed in your inbox over the weekend. You have a business to run. You can't spend 12 hours reading cover letters, many of which say almost nothing useful, and you need someone shortlisted by Friday so interviews can happen next week. Meanwhile, your current team has issues too: one employee keeps missing deadlines, another asked for a raise you're not sure how to respond to, and you still haven't written the performance review that was due two weeks ago. This is the actual week of a small business owner. It's not a recruitment crisis, it's just Tuesday. AI can cut through almost every one of these problems, and you don't need to be a tech expert to use it.
Why Hiring Eats Small Business Owners Alive
Hiring is one of the highest-stakes tasks a small business owner does, and also one of the most time-consuming. A bad hire at the $50,000–$70,000 salary level can cost you $15,000–$30,000 in lost productivity, rehiring costs, and team disruption, that's a widely cited estimate from the Society for Human Resource Management. Yet most small businesses have no HR department, no recruiter on staff, and no standardized process. The owner writes the job ad at 11pm, skims resumes in between meetings, and makes gut-call decisions that sometimes work out and sometimes don't. AI tools, specifically ChatGPT Plus, Claude Pro, and Microsoft Copilot, can build structure and speed into every stage of this process without requiring any technical knowledge whatsoever.
The key insight is this: hiring is fundamentally a writing and analyzis task. You write a job ad, you read applications, you write interview questions, you evaluate answers, you write an offer letter. Every single one of those steps involves text, and text is exactly what modern AI tools are extraordinarily good at. When you use AI in hiring, you're not automating the decision. You're automating the paperwork around the decision so you can spend your limited time on the part that actually requires your judgment: talking to people and choosing the right one.
People management has the same structure. Annual performance reviews, difficult conversations, onboarding documents, team communication, salary negotiation responses, all of it is mostly writing. A manager who uses AI to draft a performance review isn't being lazy; they're being smart about where their attention goes. The draft gives you something to react to, edit, and make your own. That's far faster than staring at a blank page for 40 minutes, which is what most managers do. In this lesson, you'll learn exactly how to use AI across the full hiring cycle and into ongoing people management, with real prompts you can copy and adapt.
Start with what you already know
Writing Job Ads That Actually Work
Most small business job ads are either too generic ('must be a team player, detail-oriented, passionate') or accidentally intimidating ('5+ years required' for a role that a sharp 2-year candidate could do). Both problems cost you good applicants. AI helps you write job ads that are clear, specific, and calibrated to the actual role, not copied from a template that was written for a Fortune 500 company. The process takes about 10 minutes. You describe the role to ChatGPT or Claude in a conversational paragraph, and you ask it to produce a structured job ad. Then you edit it. That's the whole process.
Prompt
I run a 12-person digital marketing agency. I need to hire a client account manager. This person will be the main point of contact for 8-10 small business clients, managing their monthly social media and email campaigns. They'll join weekly client calls, write brief monthly performance reports, coordinate with our two designers and one copywriter to get deliverables done on time, and flag any client issues to me before they become problems. I want someone with 2-3 years of experience, ideally in an agency or marketing role. Salary is $52,000-$58,000. We're fully remote. Write me a job ad I can post on LinkedIn and Indeed. Keep it honest and specific, no corporate fluff.
AI Response
**Client Account Manager. Digital Marketing Agency (Remote)** **About the Role** We're a 12-person digital marketing agency looking for a Client Account Manager who likes keeping things organized, clients happy, and projects moving. You'll own the relationship with 8–10 small business clients, making sure their monthly social media and email campaigns get delivered well and on time. **What You'll Actually Do** - Be the main point of contact for your client portfolio, emails, calls, the works - Join weekly client calls and keep them focused and productive - Write clear, concise monthly performance reports (we'll show you our format) - Coordinate with our designers and copywriter to hit deadlines - Spot potential client issues early and loop in leadership before they escalate **What We're Looking For** - 2–3 years in an account management, client services, or marketing coordinator role - Strong written communicator, you can explain campaign results in plain English - organized and proactive: you follow up before you're asked - Agency or marketing agency experience is a plus, not a requirement **The Details** - Salary: $52,000–$58,000 depending on experience - Fully remote, work from anywhere in [your timezone] - Full-time, Monday–Friday **To Apply** Send your resume and two or three sentences about a client relationship you managed well, what was the situation and what did you do? We review applications on a rolling basis and aim to respond within one week.
Notice what that job ad does differently from a generic template. It tells candidates what they'll actually do on a typical week. It gives an honest salary range upfront, which filters out mismatched expectations immediately. The application question ('tell us about a client relationship you managed well') does double-duty, it screens for communication skills and gives you something concrete to evaluate before you even schedule a call. These are all small business best practices baked in by the AI because you gave it enough context to work with. Edit the output for your voice and your specific situation, but the structure is solid.
Comparing AI Tools for Hiring Tasks
| Tool | Best Hiring Use Case | Strengths | Watch Out For | Cost |
|---|---|---|---|---|
| ChatGPT Plus | Job ads, interview questions, offer letters, rejection emails | Excellent at following structured instructions; strong formatting; good at adjusting tone on request | Can produce generic output if your prompt is vague, be specific | $20/month |
| Claude Pro | Resume screening summaries, performance review drafts, sensitive HR conversations | Nuanced, thoughtful tone; handles long documents well; great for delicate feedback situations | Slightly more cautious, may hedge on definitive rankings | $20/month |
| Microsoft Copilot (in Word/Outlook) | Job description drafts inside Word, email templates in Outlook, meeting summaries in Teams | Integrated into tools you already use; no copy-pasting required | Less powerful for complex multi-step prompts; needs Microsoft 365 subscription | Included with Microsoft 365 Business, from $6/user/month |
| Google Gemini (in Workspace) | Job ads in Google Docs, interview notes in Google Meet, spreadsheet scoring in Sheets | Works inside Google Docs and Gmail; good for teams already on Google Workspace | Still maturing, outputs can be less polished than ChatGPT for HR writing | Included with Google Workspace Business, from $12/user/month |
| Notion AI | Building hiring pipelines, tracking candidate notes, onboarding wikis | Excellent for organizing a full hiring process in one place; great for team collaboration on candidates | Not a writing powerhouse, better for structure and organization than generating polished copy | $10/member/month add-on |
Screening Resumes Without Losing Your Mind
Resume screening is where most small business owners waste the most time. You read 90 resumes, and 70 of them are immediately obvious nos, wrong experience level, wrong industry, clearly didn't read the job ad. The problem is you still have to open every single one to find that out. AI can compress this process dramatically. The approach is simple: paste the text of a resume (or a batch of resume summaries) into ChatGPT or Claude along with your job criteria, and ask it to evaluate the candidate against your requirements. You are still making the final call. AI is just doing the first pass so you're only deeply reading the 15 resumes that actually warrant your attention.
There's an important ethical boundary to understand here. AI should help you organize and summarize candidate information, it should not make hiring decisions for you. That's both a legal and a human judgment issue. In several jurisdictions, including New York City, there are regulations around using automated tools in hiring decisions. The safe and smart approach: use AI to create structured summaries and flag potential strengths and gaps, then use your own judgment to decide who advances. Think of AI as a very fast, very thorough executive assistant who reads everything and prepares a briefing, you're the one who decides.
AI doesn't replace your judgment in hiring
Generating Interview Questions That Actually Reveal Something
Generic interview questions produce generic answers. 'Tell me about yourself' and 'Where do you see yourself in five years?' have been so rehearsed by candidates that they reveal almost nothing about how someone actually works. behavioral interview questions, ones that ask candidates to describe specific past situations, are far more predictive of actual job performance. The format is called STAR: Situation, Task, Action, Result. AI is very good at generating targeted behavioral questions for a specific role, and it takes about two minutes. You give it the job description and ask for role-specific questions. You pick the six or eight you like best. You now have a consistent interview framework that every candidate goes through, which also makes your evaluation more fair and defensible.
Putting It All Together: Your Monday Morning Hiring Workflow
Here's what a practical AI-assisted hiring process looks like for a small business owner. Step one: you spend 15 minutes writing a detailed description of the role in plain language and use ChatGPT Plus or Claude Pro to generate a job ad. You edit it, post it to LinkedIn, Indeed, or your local job board. Step two: when applications come in, you spend 30 minutes on a Friday doing a first-pass review, using AI to summarize any lengthy resumes and flag whether each candidate meets your core criteria. You build a shortlist of 8–12 people. Step three: you use AI to generate 10 behavioral interview questions specific to the role, pick your favorite 7, and use them consistently across all interviews.
Step four is where AI saves you significant time post-interview. After each conversation, you jot down your impressions in a few sentences and paste them into ChatGPT along with your candidate notes. Ask it to help you compare candidates across your key criteria. It won't make the decision, you will, but having a structured summary of five candidates in front of you instead of scattered notes makes the decision much clearer. Step five: once you've chosen your candidate, AI drafts the offer letter. Once you've decided on rejections, AI drafts thoughtful rejection emails that don't burn bridges. The whole process is faster, more consistent, and less stressful, and you've spent your actual brain power on the part that matters: the conversations.
The cumulative time saving is real. A typical unassisted hiring process for a small business owner, from job ad to offer letter, takes 15–25 hours spread over several weeks. With AI assistance, owners consistently report cutting that to 6–10 hours. That's a full working day returned to you, per hire. For a business where the owner's time is the scarcest resource, that's a material difference. And the quality tends to be higher too, because the process is more structured, you're asking better questions, evaluating more consistently, and communicating more professionally at every stage.
Goal: Produce a complete, ready-to-use hiring kit for one role: a polished job ad, a set of 7 behavioral interview questions, and a rejection email template, all created in under 30 minutes using AI.
1. Open ChatGPT Plus or Claude Pro (either works; free tiers will also work for this task). 2. Write a paragraph describing a role you need to hire for, or a role you've hired for recently. Include: what the person does day-to-day, who they work with, what success looks like in 90 days, salary range, and whether it's remote or on-site. 3. Paste that paragraph into the AI and ask: 'Write a job ad for this role that is specific, honest, and free of corporate jargon. Include a short application question at the end that screens for communication skills.' 4. Read the output. Edit it to match your voice, change anything that doesn't sound like you or your business. 5. In a new prompt, paste your (edited) job ad and ask: 'Generate 10 behavioral interview questions for this role. Each question should ask the candidate to describe a specific past situation. Focus on the skills most critical for success in this position.' 6. Review the questions and select your 7 favorites. Save them in a Google Doc or Notion page labeled '[Role Name]. Interview Questions.' 7. Write one more prompt: 'Write a rejection email template I can send to candidates who applied for this role but won't be moving forward. It should be warm, brief, and leave a positive impression of my business.' 8. Save all three outputs, job ad, interview questions, rejection email, in one document. You now have a reusable hiring kit for this role type. 9. Post the job ad (or share it with someone for feedback) and note which parts attracted the most questions or comments from real candidates.
How This Applies to Different Business Types
- Retail and hospitality owners: Use AI to write shift supervisor or customer service job ads with realiztic descriptions of physical demands and schedule expectations, the details that filter out mismatched candidates before interviews happen.
- Consultants and agencies: Use Claude Pro to draft nuanced rejection emails to freelancers or junior hires, the tone matters more here because your industry is small and reputation travels fast.
- Healthcare practice managers: Use ChatGPT to generate compliant, professional job ads for clinical support roles; always have a healthcare HR specializt review before posting due to licensing and regulatory requirements.
- Educators and school administrators: Use AI to create structured interview rubrics for teaching positions, consistent scoring criteria that make panel interviews more objective and defensible.
- E-commerce and product businesses: Use Notion AI to build a full hiring pipeline tracker, stages, candidate notes, decision criteria, so you can manage multiple open roles simultaneously without losing track.
- Professional services (law, accounting, financial advice): Use Copilot inside Word to draft formal offer letters and employment agreements starting from AI-generated templates, always reviewed by a solicitor or employment lawyer before sending.
- Trades and construction businesses: Use ChatGPT to write job ads that lead with pay rate, hours, and physical requirements, the practical details that skilled tradespeople actually read first.
Writing Performance Reviews With AI Assistance
Performance reviews are dreaded by managers and employees equally, managers because they take forever to write, employees because they're often vague and unhelpful. AI solves the manager side of this problem entirely. The process: you spend 10 minutes jotting down bullet points about an employee's actual performance over the review period, what they did well, where they struggled, specific examples you remember. You paste those notes into ChatGPT or Claude and ask it to write a structured performance review. The AI turns your rough notes into clear, professional prose. You read it, edit it for accuracy and tone, and add anything personal that the AI couldn't know. What used to take 90 minutes now takes 25.
Prompt
I need to write a mid-year performance review for one of my employees, Jamie, who works as a customer service coordinator at my 8-person e-commerce business. Here are my notes: Jamie has been really strong on response time, our average reply time dropped from 6 hours to 2.5 hours since she joined. She handles difficult customers well and rarely escalates to me unless it's genuinely serious. She's struggled a bit with our new inventory system, took longer than expected to get comfortable with it and made a few data entry errors in April. She's improved since then. She's proactive about suggesting process improvements, she redesigned our returns tracking spreadsheet which saved the team time. She's a positive presence in team meetings. I want the review to be honest, constructive, and forward-looking. She's a keeper. I want her to feel valued but also clear on the one development area.
AI Response
**Mid-Year Performance Review. Customer Service Coordinator** **Employee: Jamie | Review Period: January–June | Prepared by: [Manager Name]** **Overall Summary** Jamie has had a strong first half of the year. She's made a measurable impact on our customer experience and brought genuine initiative to process improvement. One development area remains, and it's one she's already shown the ability to address. **Strengths** *Customer Response and Resolution* Jamie's impact on our response times has been significant. Since she joined, our average reply time has dropped from 6 hours to 2.5 hours, a 58% improvement that directly affects customer satisfaction. More importantly, the quality of those responses is high. She handles difficult customers with composure and good judgment, escalating only when genuinely necessary. This saves management time and builds customer trust. *Initiative and Process Thinking* Jamie doesn't just complete tasks, she improves them. Her redesign of our returns tracking spreadsheet is a good example: she identified a friction point, proposed a solution, and implemented it in a way that saved the whole team time. This kind of proactive thinking is exactly what a growing business needs from a coordinator-level role. *Team Contribution* Jamie is a consistent, positive presence in team meetings. She contributes constructively and supports a collaborative working environment. **Development Area** *Systems Proficiency* The transition to our new inventory system was a challenge for Jamie in Q1. There were some data entry errors in April that required correction. To her credit, she's addressed this, accuracy has improved noticeably since then. The focus for H2 is consolidating that improvement and building confidence with the system so it becomes a strength rather than a neutral. **Goals for H2** - Maintain sub-3-hour average response time as volume grows - Zero data entry errors over a 60-day period by end of Q3 - Identify one additional process improvement to propose to the team **Summary** Jamie is a valued member of this team. Her customer-facing skills and initiative make her someone we want to grow with the business. I'm looking forward to seeing her continue to build on a strong start.
Key Takeaways from Part 1
- Hiring is primarily a writing and analyzis task, and AI tools are built for exactly that. You can use ChatGPT Plus, Claude Pro, or Copilot without any technical knowledge.
- A specific, honest job ad attracts better candidates and filters out poor-fit applicants before interviews happen. AI can write that ad in 10 minutes if you give it enough context.
- Resume screening with AI means you read deeply only the resumes that deserve it. AI does the first pass and flags candidates against your criteria.
- behavioral interview questions (STAR format) are more predictive than generic questions. AI generates role-specific versions in two minutes.
- AI-assisted hiring can cut the time from job posting to offer letter from 15–25 hours down to 6–10 hours, without reducing quality. Often it improves it.
- Performance reviews become much faster when you give AI your rough notes and ask it to produce a draft. You edit for accuracy and tone, the hard thinking is already done.
- AI should assist your hiring judgment, not replace it. Final decisions are always yours. Be aware of bias risks and legal considerations around automated hiring tools.
- Different tools suit different tasks: ChatGPT Plus and Claude Pro for writing, Notion AI for pipeline organization, Copilot or Gemini if you're already inside Microsoft or Google Workspace.
From Screening to Onboarding: AI Across the Full Hiring Cycle
It's Tuesday morning. You posted a job listing for a part-time bookkeeper on Friday and woke up to 47 applications. You have three client meetings today, payroll to approve, and a supplier call at noon. Somewhere in that inbox is the right person for your business, but finding them means reading through cover letters that all say roughly the same thing, spotting the one who actually knows QuickBooks versus the one who just listed it, and scheduling interviews around everyone's availability. This is the reality of hiring as a small business owner. You're the CEO, the HR department, and the interviewer all at once. AI doesn't eliminate that responsibility, but it compresses the time it takes to do it well.
Writing Job Descriptions That Attract the Right People
Most small business job postings are either too vague or accidentally off-putting. Generic phrases like 'must be a team player' and 'fast-paced environment' signal nothing useful to serious candidates. A well-written job description does two jobs simultaneously: it sells your business to the right person and screens out the wrong ones. AI tools like ChatGPT Plus and Claude Pro are exceptionally good at this task because they've processed thousands of job descriptions across every industry. You give them the raw details, the role, the hours, the must-haves, the nice-to-haves, your business culture, and they produce a structured, compelling draft in under two minutes. The result isn't just faster; it's often more complete than what you'd write under time pressure, because the AI prompts you to think about things you'd otherwise forget to mention, like growth potential or the specific tools the person will use daily.
Prompt
Write a job description for a part-time bookkeeper at my small landscaping business. We need someone 20 hours per week, preferably mornings. They'll handle invoicing, accounts payable, payroll for 6 employees, and monthly reporting. Must know QuickBooks. Nice to have: experience with a service-based business. We're a friendly team of 8, based in Austin, TX. Salary is $22-26/hour depending on experience. Use a warm but professional tone, we're not a corporate company.
AI Response
**Part-Time Bookkeeper. Green Horizon Landscaping (Austin, TX)** We're a close-knit landscaping team of 8 looking for a detail-oriented bookkeeper to keep our finances running smoothly. If you enjoy working with a friendly crew, take pride in clean books, and want flexible morning hours, this might be your perfect fit. **What You'll Do** - Process invoices and manage accounts payable for ongoing client and supplier relationships - Run bi-weekly payroll for our team of 6 field employees - Prepare monthly financial reports for ownership review - Keep our QuickBooks file accurate and up to date **What We're Looking For** - Proven experience with QuickBooks (this is a must, please don't apply without it) - Strong attention to detail and comfort working independently - Reliable availability Monday–Friday mornings, approximately 20 hours/week - Experience in a service-based or trades business is a real plus **What We Offer** - $22–$26/hour depending on your experience - Flexible morning schedule, be done by early afternoon - A genuinely supportive team that values your work - Room to grow as our business expands To apply, send your resume and a brief note about your QuickBooks experience to [email]. We read every application and respond to all candidates.
Add Your Own Voice Before You Post
Screening Applicants Without Reading Every Word
Once applications arrive, the challenge shifts to speed and fairness. Reading 47 cover letters carefully takes two to three hours, time most small business owners simply don't have. The practical approach is to use AI as a first-pass filter. You can paste a batch of cover letters or resume summaries into ChatGPT or Claude and ask it to identify which candidates meet your must-have criteria and flag any standout details. This isn't the AI making your hiring decision. It's the AI doing what a good assistant would do: sorting the pile and surfacing the ones worth your attention. You still read the shortlisted applications. You still make the call. But you're starting from five strong candidates instead of wading through 47 mixed ones. For legal and ethical reasons, always ensure your screening criteria focus on skills and experience, not characteristics that could introduce bias.
AI Screening Has Real Legal Risks. Know Them
Preparing for Interviews Like a Pro
Most small business owners walk into interviews with a rough mental list of questions and improvise from there. That works fine for straightforward roles, but for positions where the hire really matters, a first employee, a sales lead, an operations manager, improvised interviews often miss the things you most need to know. AI can generate structured interview question sets in under a minute. More usefully, it can build behavioral interview questions tailored to the specific challenges of your role. Behavioral questions ('Tell me about a time when...') are consistently better predictors of job performance than hypothetical ones ('What would you do if...'). Ask Claude or ChatGPT to write five behavioral questions for your role, and you'll have a more rigorous interview than most small businesses run, without spending an hour thinking them up yourself.
Choosing the Right AI Tool for Hiring Tasks
| Task | Best Tool | Why It Works | Cost |
|---|---|---|---|
| Writing job descriptions | ChatGPT Plus or Claude Pro | Strong at structured writing with a specific tone; handles length and formatting well | $20/month each |
| Screening cover letters and resumes | Claude Pro | Handles longer documents; can process multiple applications in one prompt | $20/month |
| Generating interview questions | ChatGPT Plus or Claude Pro | Both excel at role-specific behavioral questions; either works well | $20/month each |
| Drafting offer letters and onboarding docs | Microsoft Copilot (Word) | Works inside Word; ideal if you already have templates to build from | Included in M365 Business at $6+/user/month |
| Creating employee handbooks | ChatGPT Plus | Good at long-form structured documents; use as a draft starting point | $20/month |
| Summarising performance notes for reviews | Notion AI | Keeps everything in one workspace; can pull from meeting notes and task history | $10/user/month add-on |
| Scheduling and interview coordination | Google Gemini with Calendar | Integrates with Google Workspace; can suggest times and draft scheduling emails | Included in Google Workspace |
| Writing rejection or follow-up emails | Any major AI tool | Simple writing task; even the free tiers of ChatGPT handle this well | Free tier available |
People Management After the Hire
Hiring is a sprint. Managing people is a marathon. Once someone joins your team, the ongoing work of feedback, communication, development, and performance management begins, and for small business owners, this is often where things slip. Annual reviews happen late or not at all. Difficult conversations get postponed. Recognition is inconsistent. AI won't fix a management culture, but it removes the friction that makes good management harder than it needs to be. Writing a thoughtful performance review for an employee takes most managers 45 minutes to an hour. With AI, you provide the raw notes, what went well, what didn't, specific examples you remember, and it structures them into a balanced, professional review in five minutes. You edit for accuracy and tone. The hard thinking is still yours; the drafting burden is gone.
Difficult conversations are another area where AI provides unexpected value, not by having the conversation for you, but by helping you prepare for it. If you need to address repeated tardiness, a performance issue, or a conflict between team members, you can describe the situation to Claude or ChatGPT and ask it to help you structure what you want to say. It can suggest how to open the conversation, what specific language to use, and how to frame feedback constructively. Many small business owners avoid these conversations simply because they don't know how to start them. Having a clear script, even one you'll adapt in the moment, makes it easier to actually have the conversation rather than putting it off for another month.
Recognition and team communication are smaller but consistent wins. AI can draft a weekly team update, a note celebrating a work anniversary, a message acknowledging a team member who went above and beyond on a tough project. These gestures matter for retention, and they're the kind of thing that disappears from a busy owner's to-do list when things get hectic. Spending two minutes with an AI tool to get a draft you can personalize and send takes recognition from 'intention' to 'action.' Over time, that consistency builds a workplace culture that reduces turnover, which is far more valuable than any single hire.
Goal: Use ChatGPT Plus or Claude Pro to write a job description and a set of interview questions for a current or upcoming role in your business.
1. Open ChatGPT Plus (chat.openai.com) or Claude Pro (claude.ai) and start a new conversation. 2. Think of a role you've hired for recently, are hiring for now, or expect to hire for in the next six months. It can be part-time, full-time, or a contractor position. 3. Write a prompt that includes: the job title, the key responsibilities (aim for 4-6 specific tasks), the must-have skills or experience, your business type and size, the location and hours, and the pay range if you have one. 4. Ask the AI to write a job description in a tone that matches your business culture, describe that culture briefly in your prompt (e.g., 'friendly but professional,' 'no-nonsense and practical,' 'creative and fast-moving'). 5. Read the output and mark anything that's inaccurate, missing, or doesn't sound like you. Edit those sections directly. 6. In the same conversation, ask the AI: 'Now write 5 behavioral interview questions specifically for this role, focused on the skills that matter most.' 7. Review the questions. Remove any that feel generic or irrelevant. Add one question of your own that only someone familiar with your business would think to ask. 8. Save both outputs, the job description and the question list, in a document or folder you'll actually use before the next interview. 9. If you're hiring now, post the revised job description this week and bring the question list to your next interview.
How This Applies to Your Specific Role
- Retail or hospitality owner: Use AI to write shift-supervisor job descriptions that clearly separate the role from a regular floor employee, most owners undersell these roles and attract the wrong candidates. Use AI to draft your onboarding checklist for seasonal staff.
- Service business owner (trades, cleaning, landscaping): AI can write field technician job descriptions that attract reliable people, not just available ones. Ask it to include a realiztic day-in-the-life section, candidates who read it and still apply are self-selected for the actual work.
- Consultant or agency owner: Use AI to write contractor agreements and role briefs for freelancers. Ask it to draft a project-specific onboarding email that sets expectations, access requirements, and communication norms upfront.
- E-commerce or product business owner: AI is useful for writing remote job descriptions that make your small team feel credible to candidates who might otherwise only consider larger companies. Specific, well-written postings signal professionalism.
- Studio, clinic, or practice owner (yoga, physio, dental, etc.): Use AI to draft staff communication policies, client-handling protocols, and performance review templates that reflect your specific service standards.
- Restaurant or café owner: Ask AI to help you write an employee handbook section covering your no-show policy, uniform standards, and tip-sharing rules, the things that create conflict when they're not written down clearly.
- Online educator or course creator: Use AI to write role descriptions for virtual assistants, community managers, or course support staff, roles that are hard to describe because they're new and often hybrid in nature.
Prompt
I'm a small business owner and I need to have a performance conversation with my office manager, Sarah. She's been with me for 2 years and is generally good at her job, but over the last 6 weeks she's missed 4 deadlines on invoicing, which caused two clients to call us asking where their invoice was. I've mentioned it once informally but nothing changed. I want to address it directly but I don't want to be harsh, she's a valued employee. Help me structure what to say in the conversation, including how to open it and how to set a clear expectation going forward.
AI Response
Here's a structure for your conversation with Sarah. Plan for about 20 minutes in a private setting. **Opening. Start with the relationship, not the problem** 'Sarah, I wanted to set aside some time to talk with you properly, because you're an important part of how this office runs and I want to make sure we're on the same page about something.' **State the specific issue, facts only, no interpretation** 'Over the last six weeks, I've noticed that four invoices went out late, late enough that two clients called us to ask where they were. I know I mentioned it briefly a few weeks ago, but I want to address it more directly today because it's continued.' **Invite her perspective, you might learn something** 'Before I say anything else, I'd like to hear from you. Is there something going on that's making the invoicing harder to stay on top of right now?' **Listen. Then, if no structural issue explains it, move to expectations.** 'Here's what I need going forward: invoices go out within 24 hours of a job completion, without exception. If something is going to delay that, I need a heads-up from you in advance, not after a client calls.' **Close with support, not just accountability** 'I'm not trying to put you on the spot, you do a lot well and I want this to work. Is there anything I can do to make it easier to hit that timeline consistently?' Follow up in writing within 24 hours summarising what was agreed. Keep it short and factual, this creates a record without feeling punitive.
Key Takeaways from This Section
- AI compresses hiring time at every stage, from writing job descriptions to screening applicants to preparing interview questions, without removing your judgment from the final decisions.
- Behavioral interview questions ('Tell me about a time when...') predict performance better than hypothetical ones. AI can generate a strong set for any role in under a minute.
- Different AI tools have different strengths. Claude Pro handles long documents well; Microsoft Copilot integrates with Word; Notion AI works inside your existing workspace.
- Screening assistance is powerful but carries legal risk. Always base AI-assisted screening on skills and qualifications, never on characteristics that correlate with protected groups.
- AI is a preparation tool for difficult conversations, not a replacement for having them. A clear script reduces avoidance and helps you deliver feedback more fairly.
- People management tasks, reviews, recognition notes, team updates, are small but high-value uses of AI that compound over time into a stronger workplace culture.
- The best prompt includes context: your business type, the specific role, the tone you want, and the constraints that matter. Vague prompts produce generic outputs that need heavy editing.
It's Thursday afternoon. You have three candidates coming in tomorrow for a customer service role, and you haven't had time to prepare proper interview questions. Your notes from the phone screens are scattered across two notebooks and a voice memo. You're also two weeks behind on writing a performance review for your longest-serving employee, the one who actually deserves a thoughtful, specific review, not a rushed paragraph. This is the exact situation where AI earns its place in your week. Not as a replacement for your judgment, but as the assistant that does the prep work so you can show up ready.
Turning Scattered Notes Into Structured Hiring Decisions
Most small business owners don't lose good candidates because they made bad decisions, they lose them because the process felt disorganised. A candidate who gets vague questions, a slow follow-up, or a generic offer letter assumes your business runs the same way. AI helps you build consistency without hiring an HR manager. You can paste rough notes into ChatGPT or Claude and ask it to generate structured interview scorecards, role-specific behavioral questions, or a candidate comparison summary. The AI doesn't know your candidates, you do. But it can turn your instincts into a format that holds up when you're comparing three people a week later and trying to remember why you preferred one over another.
Performance reviews are a similar story. Most managers, in businesses of any size, write reviews under pressure and default to vague praise or vague criticism. Neither helps the employee grow, and neither protects you if a difficult conversation comes later. AI can help you structure a review around specific examples, suggest language that is direct without being harsh, and flag where your draft is too general to be useful. You provide the facts: what the employee did, what worked, what didn't. The AI helps you say it clearly. That's the division of labor that works.
There's a real legal and ethical dimension here that small business owners must take seriously. AI-generated job descriptions, interview questions, and review language can carry unintentional bias, particularly around gender-coded language in job ads, or questions that inadvertently touch on protected characteristics. Tools like Textio and ToneAnalyzer are built specifically to catch this in hiring content. For general-purpose AI tools like ChatGPT or Claude, you need to review outputs carefully and ask explicitly: 'Check this job description for language that could discourage any group from applying.' That one extra prompt can prevent a real problem.
One Prompt That Saves Your Friday Afternoon
What AI Actually Produces: Real Hiring Outputs
Prompt
I'm hiring a part-time bookkeeper for my retail business. They'll handle invoicing, reconcile accounts weekly, liaise with my accountant, and flag cash flow issues early. I need 6 behavioral interview questions with a simple 1-3 scoring guide for each. The role requires someone who is detail-oriented, proactive about flagging problems, and comfortable working independently.
AI Response
**behavioral Interview Scorecard. Part-Time Bookkeeper** **Q1. Tell me about a time you caught a financial error before it became a bigger problem.** 1 = Vague answer, no specific example | 2 = Clear example, reactive catch | 3 = Proactive system or habit that prevented recurring errors **Q2. Describe how you organize your work when you're managing multiple deadlines without daily supervision.** 1 = Relies heavily on reminders from others | 2 = Has a basic system, some gaps | 3 = Clear, self-directed workflow with evidence it works **Q3. Walk me through how you'd flag a cash flow concern to a business owner who isn't financially trained.** 1 = Technical language, no adjustment for audience | 2 = Simplifies somewhat, but unclear | 3 = Translates numbers into plain business impact with a suggested next step **Q4. Tell me about a time you had to learn a new software tool quickly for a finance role.** 1 = Avoided or resisted | 2 = Learned it but slowly | 3 = Self-taught efficiently, can teach others **Q5. Describe a situation where you disagreed with how something was being recorded or categorized. What did you do?** 1 = Said nothing | 2 = Raised it but dropped it | 3 = Raised it clearly, proposed a fix, documented the outcome **Q6. How do you ensure accuracy when you're doing repetitive tasks like weekly reconciliations?** 1 = No clear method | 2 = Checks work occasionally | 3 = Consistent checklist or verification habit with zero-error track record
Comparing AI Tools for Hiring and People Tasks
| Tool | Best For | Hiring Use Case | Review Use Case | Cost |
|---|---|---|---|---|
| ChatGPT (Free/Plus) | Interview prep, job ads, offer letters | Question generation, scorecard drafting | Review structure, feedback language | Free / $20/month |
| Claude (Free/Pro) | Long-form drafts, nuanced feedback | Full hiring workflow from JD to offer | Detailed performance narratives | Free / $20/month |
| Microsoft Copilot | Teams/Outlook-integrated HR tasks | Summarising candidate email threads | Review drafts inside Word | Included in M365 Business |
| Textio | Bias-free job descriptions | Language analyzis for inclusive JDs | Review language equity scoring | Paid, enterprise pricing |
| Notion AI | centralizing hiring docs and notes | Candidate tracking, template library | Review storage and version history | Included in Notion plans |
Putting It Into Practice This Week
Start with the highest-friction task on your people management list right now. If you're hiring, that's probably your job description or your interview prep. If you're managing, it's likely a performance conversation you've been postponing because you don't know how to frame it. Pick one. Open Claude or ChatGPT, describe the situation in plain language, no special format required, and ask for a draft, a structure, or a set of questions. Your first output won't be perfect. It rarely is. But it will give you something to react to, and reacting is faster than starting from nothing.
The second habit worth building is saving your prompts. Every time AI produces something useful, a job description, a set of interview questions, a review template, save the prompt alongside the output in a simple document or Notion page. Within three months, you'll have a reusable hiring toolkit that reflects your business, your language, and your standards. You won't start from scratch the next time you hire. That compounding value is where small business owners get a real structural advantage from AI, not from any single output, but from the library they build over time.
Finally, keep humans in the loop on every consequential decision. AI helps you prepare, structure, and communicate, but you decide who gets hired, who gets promoted, and who has a difficult conversation. A performance improvement plan drafted with AI still needs your signature, your judgment, and your relationship with that employee. The legal, ethical, and human dimensions of people management don't get outsourced. They get supported. That's the right frame for every AI tool in this space.
Goal: Create a reusable set of hiring documents for one role in your business using free AI tools, ready to use the next time you recruit.
1. Choose one role you hire for regularly or are currently recruiting for. Write 3-4 sentences describing what the person actually does day-to-day. 2. Open ChatGPT (free) or Claude (free) and paste your description. Ask it to write a job advertisement under 250 words that would attract strong candidates and discourage poor fits. 3. Read the output. Edit anything that doesn't sound like your business. Add one specific detail only you would know, a real aspect of the role or your culture. 4. In the same chat, ask: 'Now generate 6 behavioral interview questions for this role with a 1-3 scoring guide for each answer.' 5. Copy the scorecard into a Google Doc or Word document. Label it with the role name and today's date. 6. Ask the AI: 'Write a short rejection email and a short offer email template for this role, professional but warm in tone.' 7. Save both email templates in the same document. You now have a three-part hiring toolkit: job ad, scorecard, and email templates. 8. Paste your job description back into the chat and ask: 'Check this for any language that could unintentionally discourage women, older candidates, or any other group from applying.' Review and act on any suggestions. 9. Save the final document somewhere you'll find it, a shared drive, Notion, or a clearly labeled folder. Next time you hire for this role, open it before you post anything.
How This Applies Across Different Roles
- Retail owner: Use AI to write shift supervisor job ads, generate customer-facing role scorecards, and draft 90-day check-in templates for new hires.
- Consultant or agency owner: Ask AI to help you write subcontractor briefs, structure freelancer performance check-ins, and draft project-end feedback summaries.
- Café or hospitality owner: Generate onboarding checklists, write kitchen or front-of-house role descriptions, and draft probation review templates in plain language.
- Trades or services business: Use AI to write apprentice or technician job ads, build skills-based interview questions, and create simple performance trackers.
- Online or e-commerce business: Ask AI to draft virtual assistant role descriptions, remote worker onboarding guides, and async feedback frameworks for distributed teams.
- Teacher or education provider: Use AI to write tutor or classroom aide job descriptions, structure reference check questions, and draft staff feedback that meets professional standards.
Prompt
I need to have a performance conversation with a long-term employee who has become unreliable with deadlines over the past two months. They're a good person and have been with me for four years, but this is affecting the team. I don't want to be harsh but I need to be direct. Help me structure the conversation and give me the key phrases I should use.
AI Response
**Performance Conversation Framework. Deadline Reliability Issue** **Opening (set the tone, not punitive, not vague)** 'I wanted to talk because I value what you bring to this team and I want to make sure we're set up for things to keep going well. I've noticed something over the last couple of months that I think we need to address together.' **State the specific pattern (not character, not personality)** 'Over the past eight weeks, I've noticed that [specific project] came in three days late, [task] missed the Friday deadline twice, and I've had to follow up more than usual to get updates. I want to understand what's going on.' **Listen before you solve** Pause here. Ask: 'Is there something happening that I should know about?' Their answer changes your next step entirely. **Be clear about the impact** 'When deadlines slip, it creates pressure on the rest of the team and affects our ability to deliver for clients. I need this to change.' **Agree on a specific, measurable next step** 'What I'd like us to agree on today is [specific expectation, e.g., weekly check-in every Monday, deadlines confirmed 48 hours in advance]. Does that work for you?' **Close with clarity, not ambiguity** 'I'm not raising this to put you under pressure. I'm raising it because I want this to work. Let's check in again in three weeks and see how it's going.' *Document the conversation date and the agreed next step in writing after the meeting.*
Key Takeaways
- AI handles the prep work in hiring, job ads, interview questions, scorecards, offer letters, so you can focus your energy on the actual decision.
- Performance reviews and difficult conversations become easier when AI helps you structure your thoughts and translate vague impressions into specific, useful language.
- Always review AI-generated hiring content for bias, especially in job descriptions. One extra prompt asking for a bias check takes 30 seconds and can prevent a real problem.
- Save every useful prompt alongside its output. Within months, you'll have a reusable people management toolkit built around your business.
- The right tools depend on your workflow: ChatGPT and Claude for drafting, Textio for bias-checking, Microsoft Copilot if you're in the Microsoft ecosystem, Notion AI for organizing and storing.
- AI supports your judgment, it does not replace it. Every hire, review, and difficult conversation still requires your human oversight, relationship awareness, and final call.
Sign in to track your progress.