This week we’re talking about two things that sound very ordinary - **spring cleaning** and **giving feedback to a team member** - and looking at how AI can quietly make both easier, faster, and less stressful. One is about getting your physical space under control without getting overwhelmed. The other is about handling a human conversation you might be dreading, and doing it with more clarity and kindness.

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

On a rainy Saturday in early July, Denise, 58, is standing in her kitchen staring at a sticky note that just says: “CLEAN HOUSE.” She’s had a chaotic six months - aging parents, grown kids boomeranging home, work deadlines - and every room shows it. She keeps opening cabinet doors and thinking, “I should deal with this,” then closing them again.

The problem isn’t effort. It’s overwhelm. There’s too much to remember: the dusty ceiling fans, the fridge that probably has something growing in the back, the pile of shoes at the front door. She doesn’t want another generic “spring cleaning checklist” that assumes she lives in a 5‑bedroom show home. She wants something that fits *her* actual house, *her* energy level, *this* weekend.

So instead of staring at the sticky note, she opens her AI assistant and has it build a tailored checklist she can realistically finish in two afternoons, with boxes to tick and breaks built in. It feels less like chaos and more like a plan.

Here is the exact prompt to use:

Copy and paste this into ChatGPT or Claude:

> I want you to act as a practical home-organizing coach and create a realistic spring cleaning checklist tailored to my actual home and limited energy. > > Ask me up to 10 quick questions first about my home: number of rooms, whether I have kids or pets, how cluttered things are right now, and how much time and energy I honestly have this week. Use my answers to build a **two-day spring cleaning plan** that: > - Breaks tasks into short, specific actions (for example: “Empty and wipe one kitchen drawer” rather than “organize kitchen”). > - Groups tasks by room and in the order I should do them so I don’t need to think about where to start. > - Starts with decluttering surfaces and obvious trash before deeper cleaning, following the idea that clutter should be handled before scrubbing. > - Includes 20-minute cleaning blocks followed by 10-minute breaks so I don’t burn out. > - Adds 3 - 5 “often forgotten” tasks for each room (like baseboards, ceiling fans, filters, under furniture). > > When you present the plan, give me: > - A short overview of the two days. > - Room-by-room checklists with easy-to-tick items. > - A suggested order for the tasks (what to do first, second, third). > - One sentence at the end reminding me that it’s okay if I don’t finish everything. > > At the end, ask if I want to adjust anything (for example: shorter sessions, different rooms, or moving tasks to next week) and then revise the checklist once.

Why this prompt works:

You’re asking the AI to **interview you first**, which keeps it from giving you a generic, unrealistic list. You specify time limits, energy level, and structure (20/10 blocks, room-by-room order), so you get something you can actually follow without thinking. You also explicitly request “often forgotten” tasks and a gentle tone, which nudges the AI away from perfectionism and toward something more humane and practical.

AI doesn’t know your fridge is a disaster or that your back starts hurting after 30 minutes unless you *tell* it. If the first checklist feels too ambitious, say so and ask it to cut the list in half, or to focus only on two rooms this week. Also, AI can miss unusual items (like a big aquarium or a home gym); scan the list and add or remove tasks so it matches your real space.

Use Case 2 of 2

Use Case 2: Giving feedback to a team member

On Wednesday afternoon, Marcus, 52, is sitting in his car outside the office, rehearsing a conversation he doesn’t want to have. One of his team members - Lena, 34 - has been missing deadlines and handing in work that’s “almost there” but not quite. They’ve worked together for years. He likes her. He also knows that if he doesn’t address this, the rest of the team will quietly resent doing cleanup on her projects.

Marcus is not worried about what’s *true*. He knows the facts: three late reports in two months, clients asking for revisions, other colleagues stepping in. He’s worried about how it will *land*. He doesn’t want Lena to feel attacked or blindsided. He wants to be specific, fair, and kind. The trouble is, when he sits down to write out what he wants to say, the words either sound too harsh or so watered down that they don’t say anything at all.

Instead of winging it, he decides to treat the AI like a drafting partner. He gives it the situation, asks for a script he can say out loud in his own voice, and uses it to shape a conversation that is clear but respectful. He still owns the feedback and the decision; he just doesn’t have to invent every sentence from scratch.

Here is the exact prompt to use:

Copy and paste this into ChatGPT or Claude:

> I have to give performance feedback to a team member, and I want your help drafting what to say. > > Here are the details (do not include this section in your final answer): > - My role and level of authority: [briefly describe]. > - Their role and how long we’ve worked together: [briefly describe]. > - The specific issues I’ve observed, with 2 - 3 concrete examples and dates: [list them clearly]. > - Any strengths or positives I genuinely see in their work: [list at least 2]. > - What outcome I want: for example “clear improvement over the next 30 days,” “agreement on new deadlines,” or “honest discussion about whether this role still fits.” > - Any constraints: HR policies, remote worker, sensitive personal situation, etc. > > Based on this, write: > 1) A short opening I can use in a one-on-one meeting that sets a calm, respectful tone. > 2) A clear, specific explanation of the issues, using “I” statements and focusing on behaviors and impact, not personality. > 3) A balanced section that acknowledges their strengths and past contributions without minimizing the problem. > 4) 3 - 4 questions I can ask them to invite their perspective and ideas for improvement. > 5) A closing that summarizes next steps and, if appropriate, expresses my hope that we can solve this together. > > Keep the language direct, professional, and human - I’m in my 40s/50s and want it to sound like me, not like a corporate memo. Avoid buzzwords and overly emotional language. > > At the end, give me a short checklist of key points to cover so I can adapt the script in my own words.

Why this prompt works:

You’re feeding the AI **specific context and examples**, which is what makes feedback sound grounded rather than vague or accusatory. You clearly outline the structure you want (opening, explanation, strengths, questions, closing), so the AI organizes the conversation for you instead of producing a random block of text. By asking for a checklist at the end, you give yourself a way to use the ideas without feeling like you have to read a script word for word.

AI cannot know your workplace culture, Lena’s personal history, or the emotional dynamics in the room. Review the draft and cross out anything that feels unlike you, too stiff, or unintentionally sharp. Also, treat the AI’s version as a *starting point*, not a verdict - you are still responsible for the fairness of the feedback and for checking it against HR guidelines or any legal requirements in your organization.