This week we’ve got two very down‑to‑earth uses for AI: **turning spring cleaning into a painless checklist** and **giving thoughtful feedback to a team member without tying yourself in knots**. Both matter because they tackle the kind of mental load that creeps up on you: clutter in your home, and tension in your working relationships. AI can’t scrub your oven or hold the performance review for you - but it can make both jobs much lighter.

Try at least one before Sunday. That is the whole assignment.

Use Case 1 of 2

Use Case 1: Spring cleaning using a checklist AI creates

Let’s start with Claire. She’s 57, lives in a three‑bedroom townhouse, and every April she promises herself she’ll “really get on top of the house this year.” By mid‑June, she has half‑cleaned closets, three donation bags sitting by the front door, and a vague sense of defeat every time she walks past the guest room.

Last Saturday morning, around 9:30, she opened her notebook and tried to make a list. It quickly turned into a jumble: “windows? pantry? go through kids’ old school papers? garden furniture? figure out garage.” She stared at the page, got overwhelmed, and decided to “start with the kitchen,” which then took the entire day because she had no sense of what was realistic.

What Claire really needed wasn’t motivation - it was **structure**. A clear checklist, broken down by room, with reasonable chunks she could spread across a few weekends. She didn’t need an app with 40 features. She needed someone to think it through with her. That’s exactly the kind of thinking AI is good at: turning a vague goal (“spring clean the house”) into a practical sequence of actions that fit your life.

Here is the exact prompt to use:

Copy and paste this into ChatGPT or Claude:

> I want to do a realistic spring cleaning of my home, and I need you to turn this into a clear, step‑by‑step checklist. > > Here are the details about my home and my situation: > - Type of home: [describe: e.g., 3‑bedroom townhouse, 2 bathrooms, finished basement, small yard] > - People living here: [e.g., me and my partner, two adult children visit occasionally] > - Biggest pain points: [e.g., cluttered closets, messy pantry, garage chaos, dusty blinds, papers everywhere] > - My available time: [e.g., 2 hours on weeknights, 4 hours on Saturdays, none on Sundays] > - My energy level: [e.g., I get tired easily and can’t do heavy physical work for more than 1 hour at a time] > - Any health or mobility limits: [e.g., bad knees, can’t do ladders] > - What “success” looks like for me: [e.g., surfaces clear, closets usable, kitchen and bathroom feel fresh, garage not perfect but I can find things] > > Based on this, please: > 1. Create a **room‑by‑room spring cleaning checklist** with specific tasks for each space. > 2. Group tasks into **manageable sessions** (no more than 2 hours per session) that I could realistically do over 4 - 6 weeks. > 3. Mark each task as **“essential”** or **“optional”** so I can skip things if I run out of time. > 4. Suggest a **simple schedule** (week by week) that spreads the work out and avoids overloading me. > 5. For any task that’s vague (like “declutter closet”), spell it out into clear steps a non‑organizer could follow. > > Use plain language, no jargon. Assume I’m not very tech‑savvy and prefer to print a list and check things off.

Why this prompt works:

You’re giving the AI **context** (type of home, your time, your energy), a **clear outcome** (“room‑by‑room checklist over 4 - 6 weeks”), and **constraints** (session length, essential vs optional). That nudges it to design something humane instead of a perfectionist marathon. Asking it to spell out vague tasks (“declutter closet”) forces practical detail instead of airy advice, and specifying plain language keeps it usable.

AI will often be more ambitious than you are. It may suggest you can do four rooms in a weekend, or add tasks that feel fussy (like wiping down door hinges) when you’re just trying to get out from under the mess. Treat the checklist as a **draft**. Cross out anything that makes you groan, and adjust the schedule to fit your real life. The goal isn’t to impress the AI; it’s to create a plan you’ll actually follow.

Use Case 2 of 2

Use Case 2: Giving feedback to a team member

Now, meet David. He’s 49, leads a small project team, and has a recurring knot in his stomach about one person: Elena, a smart analyst who does solid work but has a habit of missing internal deadlines. The external deliverables go out on time, but David ends up doing last‑minute triage and quietly fixing her slides.

Last Thursday, around 3 p.m., he got yet another “running a bit behind, will send tonight” message from her. He sat at his desk, opened a blank email, and stared at the screen. He didn’t want to sound angry. He didn’t want to write one of those vague “We should align on expectations” notes that mean nothing. He also didn’t want to wait for her annual review and dump twelve months of frustration on her in one sitting.

This is exactly where AI can help: not by deciding what he *feels*, but by helping him **turn his muddled frustration into clear, fair, specific feedback**. Think of it as a drafting partner - something that can outline the email or talking points so he can focus on the substance, not the phrasing.

Here is the exact prompt to use:

Copy and paste this into ChatGPT or Claude:

> I need help drafting professional, constructive feedback for a team member. I don’t want you to decide what I feel or what I should do, but I do want help turning my thoughts into clear, balanced wording. > > Here is the situation in plain terms: > - My role: [e.g., project lead in a mid‑size consulting firm] > - Their role: [e.g., analyst responsible for research and slide preparation] > - What they do well: [list 3 - 4 concrete strengths, e.g., “good research quality,” “asks useful questions,” “clients like her style”] > - The specific problem: [describe 2 - 3 recent examples with dates, what was expected, what actually happened, and the impact on you / the team] > - What I want to change: [e.g., “I need her to meet internal deadlines or tell me early if she can’t”] > - My tone: [e.g., calm and direct, not angry, no sugar‑coating but also no harsh language] > - How I plan to deliver it: [e.g., 1:1 conversation next Tuesday, followed by a short email recap] > > Based on this, please: > 1. Draft a **short feedback email** I can send after our conversation, summarizing the key points. > 2. Suggest **3 - 4 talking points** I can use in the actual conversation, in simple, direct language. > 3. Make sure the feedback is **specific**, focused on behaviors and impact, not on personality. > 4. Include one or two **questions I can ask** her to invite her perspective (e.g., about what’s blocking her from meeting deadlines). > 5. Keep everything in a style that would feel natural for a manager in their 40s or 50s, not full of corporate buzzwords. > > After you draft this, show me: > - One version that is **very direct**, and > - One version that is **a bit softer**, > so I can choose which fits me better.

Why this prompt works:

You’re feeding the AI **specific examples**, not vague “she’s not reliable” complaints. That lets it frame feedback around behaviors and consequences, which comes across as fair instead of personal. You also separate the **conversation** (talking points) from the **written record** (follow‑up email), which are slightly different animals. Asking for two versions (direct and softer) gives you options without you having to wordsmith from scratch. The instruction “don’t decide what I feel” keeps the AI in its lane: drafting, not therapy.

AI‑generated feedback can sound a bit generic or overly polished - like it was pulled from a management textbook. Before you use it, **read it out loud** and tweak any phrases you’d never actually say. Also, be careful not to hide behind the email. The real work is the conversation where she can respond and you can listen. Use AI to get your thoughts in order, not to avoid an honest discussion.