This week, we’re talking about **spring cleaning with an AI-made checklist** and **using AI to give better feedback to a team member**. On the surface these sound small and domestic, but they touch two big themes: how you manage your environment and how you manage your relationships at work. Both are places where tiny improvements compound over time.

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 Tuesday evening, around 7:15 p.m., Karen dropped onto her living room sofa and did that slow 360° glance many of us know too well. The coffee table had mail from three weeks ago. The hallway baseboards had that gray dust stripe. Her guest room had quietly become a “temporary” storage zone sometime in 2024 and never recovered.

She wasn’t lazy; she was tired. Karen works full-time, helps her mom with medical appointments, and she’s the default “organizer” for her extended family. The idea of “spring cleaning” felt laughable. She didn’t need another Pinterest-perfect checklist; she needed something realistic that matched her actual house and actual energy.

A friend mentioned using AI to build a cleaning plan. Karen was skeptical, but the idea of “just tell it my rooms and my schedule, and it turns that into a checklist I can follow” sounded… appealing. Not a miracle. Just one less thing to plan. She opened her AI app and decided to try it, not to become a different person, but to make it slightly easier to be the person she already is.

Here is the exact prompt to use:

Copy and paste this into ChatGPT or Claude:

Copy and paste this into ChatGPT, Claude, Gemini, or any similar tool, then customize the parts in brackets to fit your life:

> I need a realistic spring-cleaning checklist for my home. > > My home has these areas: [list your rooms and spaces - e.g., kitchen, living room, hallway, 2 bathrooms, 3 bedrooms, home office, garage, front porch]. > > I have about [X] hours per day and [Y] days total to work on this. I prefer to work in short blocks, not marathon sessions. > > Please create a plan that: > - Starts with the most visible and highest-impact areas first (the places guests would notice) > - Is broken into day-by-day tasks I can do in 20 - 30 minute blocks > - Includes a short decluttering step before deep cleaning in each room > - Gives me a room-by-room checklist with simple language > - Includes a list of cleaning supplies I should gather first > - Uses a simple “keep, donate, throw away” decision rule for clutter > - Ends with a final day for finishing touches and commonly forgotten spots like baseboards, vents, light switches, door handles, and filters > > I am [describe your situation - e.g., “working full-time with mild knee pain, so I can’t be up and down on ladders” or “sharing the home with two teenagers who can help with some tasks”]. Adjust the plan to fit my energy and limitations. > > Keep the checklist clear and not too long. If a task is optional, please label it as OPTIONAL so I can skip it without guilt.

Why this prompt works:

You’re not asking for a generic cleaning list; you’re giving the AI a **clear picture of your home, time, and constraints**, then telling it how to structure the output: short blocks, visible areas first, declutter before deep clean, and a final “forgotten spots” pass. That combination gives you something you can actually follow instead of an aspirational chore novel. Labeling OPTIONAL tasks upfront helps you avoid perfectionism and decision fatigue.

AI doesn’t know your actual physical limits or safety issues. If it suggests climbing on chairs, moving heavy furniture, or using harsh chemicals, use your judgment and skip anything that feels risky. Also, the checklist might still be too ambitious the first time; treat it as a draft. Ask the AI to “cut this list in half and make it more realistic for someone who’s already tired” if your first reaction is overwhelm. You’re allowed to negotiate with it.

Use Case 2 of 2

Use Case 2: Giving feedback to a team member

On Wednesday at 3:40 p.m., right after a tense Zoom call, David stared at the draft email on his screen and felt that familiar knot in his stomach. He manages a small team of six at a nonprofit. One of his staff members, Lila, had missed a key deadline - again - and her work lately had been rushed and full of small mistakes.

The board was starting to notice, and it was landing on David. He knew he had to address it directly, but Lila wasn’t a villain; she was overwhelmed, new to the role, and genuinely trying. David didn’t want to send a harsh email that crushed her. He also didn’t want another vague, polite message that avoided the issue and changed nothing.

He’d heard that AI could help “rewrite” emails, but he wasn’t looking for fluff. He wanted: - Clear, specific feedback - Respectful tone - A path forward, not just criticism

So instead of asking AI, “Write an email to my team member,” he pasted his own rough draft - the blunt, slightly frustrated version he’d typed - and asked the AI to reshape it while keeping the hard truths.

Here is the exact prompt to use:

Copy and paste this into ChatGPT or Claude:

First, write your own honest, rough draft of the feedback you want to give (even if it’s too blunt to send). Then copy that draft and paste it into your AI tool with this prompt:

> I am a manager giving performance feedback to a team member. Below is my rough draft email. > > Please rewrite it so that: > - The feedback is **clear and specific** about the behavior and impact > - The tone is **respectful, direct, and calm**, not harsh or vague > - It focuses on **observable actions** (missed deadlines, errors, communication gaps) rather than personal judgments > - It includes **concrete next steps** and support I can offer > - It invites the team member to respond or add context > > Keep it in my voice as much as possible and do not make it overly formal or overly cheerful. > > After you rewrite it, list: > 1) The key messages you think I’m trying to convey > 2) Any parts that might still feel too harsh or too soft > 3) One suggestion for a sentence I could say out loud if we discuss this in a meeting instead of just email > > Here is my rough draft: > [paste your actual draft email here]

Why this prompt works:

You’re not handing the whole situation over to the AI; you’re starting with your own words and using the AI as a **clarity and tone editor**, not a ghostwriter. The instructions force it to keep the feedback specific and tied to actions, which is what makes feedback fair and useful. Asking for a short analysis at the end (“key messages,” “too harsh or too soft”) helps you see where the AI may have softened something too much or dodged a real issue.

AI tools often default to overly nice, somewhat vague language - lots of “I really appreciate…” and not enough “We need to change X by Y date.” Read the revised email critically and put back any necessary firmness the AI diluted. Also, privacy matters: don’t paste sensitive HR details or medical information; keep the draft focused on work behaviors and outcomes, not personal life or protected categories. Use your internal policies as the final guardrail, not the AI’s judgment.