Some weeks, “using AI” looks like writing a research report or automating some complicated workflow. This isn’t one of those weeks. Today we’re going to use AI for two very ordinary, very human problems: finally getting that spring cleaning done without melting down, and giving honest feedback to a team member without tying yourself in knots. These are small, practical wins - but they add up fast.

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 gray Saturday in early April, around 10:30 a.m., Maria stood in the middle of her living room, hands on hips, overwhelmed. She’s 52, works full-time in HR, and every year she tells herself, “This is the year I get the house under control.” The problem isn’t effort. It’s that she doesn’t know where to start. The closets are full of “I’ll deal with that later.” The garage is a maze. She starts in one room, finds something that belongs in another, wanders over there, and suddenly she’s sorting photos from 2009 while the vacuum sits in the hall.

By noon she’s exhausted, surrounded by half-finished piles, with nothing actually completed. When she tries the advice she finds online, the lists are either too generic (“declutter your home”) or written for people who apparently live in minimalist showrooms and have three possessions. None of it feels like it fits *her* house, *her* time, *her* energy. What she really needs is someone standing beside her saying: “Here’s today’s plan. Do this first. Don’t worry about the rest yet.”

That “someone” can be an AI, if you tell it enough about your reality.

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 build a realistic spring cleaning plan for me.

First, ask me clarifying questions to understand: - the size and layout of my home (number of rooms, any tricky spaces like a garage or basement) - who lives here and whether anyone will help - my physical limits (e.g., back issues, can’t do ladders, etc.) - the time I realistically have available each week - my top three priorities (for example: clear surfaces, clean floors, get rid of old stuff)

Ask the questions one group at a time and wait for my answers before continuing.

Once you have my answers, create: 1) A prioritized spring cleaning checklist broken into short, concrete tasks that can be done in 20 - 30 minutes each. 2) A simple schedule for the next four weeks, showing which tasks to do on which days, assuming I can work [X] minutes per day or [Y] hours per week. (You can ask me to fill in X and Y.) 3) A “do not worry about this yet” list of things I can safely ignore until the main priorities are done.

Make the checklist extremely specific and practical for someone who gets overwhelmed easily and tends to start too many things at once. For each task, include: - a clear, plain-language title - a very short description (2 - 3 sentences) of what “finished” looks like - what supplies I’ll need, if any

At the end, ask me: “Does this plan feel too heavy, too light, or about right?” and be ready to adjust the plan based on my feedback.

Why this prompt works: This prompt forces the AI to *ask questions first* instead of dumping a generic list on you. You tell it about your home, your limits, and your time, so the checklist is grounded in your reality, not an idealized version. Breaking tasks into 20 - 30 minute chunks, with clear definitions of “finished,” keeps you from wandering and half-completing everything. The “do not worry about this yet” list is important - it gives you permission to ignore non-essentials and focus on progress you can actually see.

Use Case 2 of 2

Use Case 2: Giving feedback to a team member

It was a Tuesday afternoon at 3:45 p.m. when David realized he’d been staring at the same email draft for almost 25 minutes. He’s 58, runs a small consulting practice, and has worked with every type of personality over the years. Right now he’s stuck on one: a capable but inconsistent team member named Jenna. She’s smart, creative, and when she’s “on,” clients love her. But she misses deadlines, makes small but important mistakes, and sometimes disappears for a day without checking in.

David has a check-in call with her at 4:30. He’s torn between two unhelpful extremes: sugar-coating things so much that nothing changes, or venting his frustration in a way he’ll regret. He wants to be fair. He wants to be kind. He also wants the work to get done at the standard he promises clients. What he doesn’t have is a script - or even a clear structure - for how to say: “This isn’t working, and it has to change” without turning it into an emotional mess.

He doesn’t need AI to *decide* what’s right or wrong. He needs help saying what he already knows, in words that are honest, specific, and calm.

Here is the exact prompt to use:

Copy and paste this into ChatGPT or Claude:

I need help preparing feedback for a real person I work with. I want you to act as a thoughtful leadership coach who respects both clarity and kindness.

First, ask me a few questions to understand: - my role and this person’s role (for example: I’m their manager, client, collaborator, etc.) - what this person does *well* that I genuinely value - the specific behaviors that are causing problems (with 2 - 3 concrete examples) - how these behaviors affect me, the team, or our clients - what I realistically need to be different over the next 30 - 60 days

Ask these questions one group at a time and wait for my answers before continuing.

Then, based on my answers, write: 1) A short summary I can say at the start of the conversation that sets the tone (2 - 3 sentences). 2) A clear feedback “script” I can use during a 20 - 30 minute conversation. The script should: - start by acknowledging what they do well - describe the problem behaviors using specific examples, not vague labels - explain why these behaviors are a problem, in practical terms - describe exactly what I need them to do differently and by when - invite their perspective and create space for them to respond 3) 2 - 3 possible phrases I can use if they react defensively, so I can stay calm but firm.

Use plain, straightforward language that would sound natural coming from a person in their 40s - 60s. Avoid corporate buzzwords like “synergy” or “leveraging.” At the end, ask me if any part feels too harsh, too vague, or unlike how I actually talk, and help me adjust the wording.

Why this prompt works: You’re not asking the AI to judge the person; you’re asking it to structure and phrase what *you* already think and feel. By feeding it real examples and clear expectations, the AI can turn your jumble of thoughts into a calm, coherent script. Having it start with questions keeps you from getting a generic “feedback sandwich” and instead creates something that sounds like you. Asking for phrases to handle defensiveness gives you a backup plan when the conversation gets uncomfortable.