On Monday morning, October 19, 2026, many of us will be looking at two familiar kinds of work: the household jobs we keep postponing and the workplace conversations we keep rehearsing in our heads. This week’s topics are using AI to create a practical spring-cleaning checklist and using AI to prepare thoughtful feedback for a team member. Neither requires technical skill. Both begin with giving the AI enough real-world detail to be useful.

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

At 8:15 a.m. on Saturday, April 18, 2026, Margaret Lewis stood in her kitchen holding a roll of rubbish bags and a cup of coffee that had already gone cold. She had promised herself she would “finally get the house sorted” that weekend, but the phrase was too large to act on. The basement needed attention. The spare bedroom had become a storage room. The windows were dusty, the linen cupboard was overflowing, and she had no idea where to begin. Margaret did not need a magazine-perfect home or a punishing weekend of cleaning. She needed a sensible plan that fit around her bad knee, her husband’s limited mobility, and the family lunch scheduled for Sunday. That is precisely the kind of planning AI can handle well - provided she describes her actual home and her actual limits.

Here is the exact prompt to use:

Copy and paste this into ChatGPT or Claude:

> I want you to create a realistic spring-cleaning checklist for my home. Do not give me a generic list. Ask me up to 10 questions first about my rooms, priorities, time available, physical limitations, pets, and any areas I want to avoid or handle carefully. After I answer, create a plan spread across four two-hour sessions over two weekends. > > Organize the plan in this order: > > 1. Tasks that will make the biggest visible difference > 2. Tasks that improve hygiene or safety > 3. Tasks that can wait > > For each task, include: > - The room or area > - A plain-English description of what to do > - An estimated time > - The supplies needed > - Whether it can be done sitting down > - Whether I should ask someone else to help > > Keep each session to no more than two hours. Include short breaks every 30 - 40 minutes. Do not recommend lifting heavy objects, climbing on furniture, mixing cleaning chemicals, or using harsh products unless you clearly explain a safer alternative. At the end, create a short shopping list and a “good enough for today” version of the plan. > > Use a calm, practical tone. The goal is a cleaner, safer, more comfortable home - not perfection.

Why this prompt works: It gives the AI a job, a process, and boundaries. Asking questions first prevents it from assuming that every home, body, and weekend is the same. The time limit makes the plan usable rather than aspirational. The request for sitting-down tasks and help with lifting turns a vague cleaning list into something adapted to real life. Most useful of all, “good enough for today” protects against the all-or-nothing thinking that causes many cleaning projects to stall.

Use Case 2 of 2

Use Case 2: Giving feedback to a team member

At 4:40 p.m. on Wednesday, October 21, 2026, Daniel Ortiz was staring at a half-written message to Priya, a capable employee whose recent reports had contained several avoidable errors. Daniel was frustrated, but he was also aware that frustration could leak into his wording. Priya had taken on extra work after a colleague left, and the mistakes were not a pattern of laziness. Still, two clients had received incorrect figures, and the team had spent hours correcting them. Daniel wanted to be clear without humiliating her, and supportive without pretending the problem did not matter. AI could not have the conversation for him. It could, however, help him separate facts from assumptions, identify the central message, and rehearse language that was direct and fair.

Here is the exact prompt to use:

Copy and paste this into ChatGPT or Claude:

> Help me prepare for a feedback conversation with a team member. Do not write a vague motivational speech and do not diagnose her personality or intentions. > > Here are the facts: > - Team member’s name: Priya > - Her role: project coordinator > - Strengths I want to acknowledge: she is responsive, dependable under pressure, and good at keeping clients informed > - Specific concern: during the last three weekly reports, two figures were copied incorrectly into the final client version > - Impact: two clients received inaccurate information, and another team member spent about four hours correcting the reports > - Context: Priya has been covering additional work since a colleague left six weeks ago > - Desired improvement: before sending a report externally, she should use the agreed checking procedure and ask for a second review when the workload makes that difficult > - My goal: preserve trust while making it clear that the errors must stop > > Please help me prepare in five parts: > > 1. Summarize the issue in one neutral sentence based only on the facts. > 2. Suggest a respectful opening that acknowledges Priya’s strengths and the extra workload. > 3. Write three versions of the main feedback: gentle, direct, and very direct. Each must describe observable behavior, its impact, and the expected change. > 4. Give me five questions that invite Priya’s perspective without sounding accusatory. > 5. Draft a short closing that confirms the next step, offers appropriate support, and sets a date for checking progress. > > Flag any sentence that sounds like a judgment about motive, attitude, or character. Replace judgmental language with specific, observable language. Keep the tone professional, humane, and concise. Do not promise consequences or support that I have not described.

Why this prompt works: It distinguishes evidence from interpretation. “Two figures were copied incorrectly” is actionable; “you are careless” is a character judgment that invites defensiveness. The prompt also supplies the impact, the desired change, and the surrounding context, so the AI can help Daniel be both accountable and fair. Asking for several levels of directness gives him options without asking the tool to decide what the relationship requires.