Smarter by Thursday logo

Smarter by Thursday

Archives
Log in
Subscribe
August 6, 2026

Issue 32: Smarter by Thursday

Smarter by Thursday — Issue 32

Issue 32 · week of August 03, 2026

Smarter by Thursday

One practical AI win, every week. No jargon required.

By Dr. Rowan Hayes · Estimated read time: 6 minutes

This week we’re tackling two situations that come up all the time once you’re past 40: **how to research a major purchase without getting overwhelmed or sold to**, and **how to write a professional “no” email that protects relationships and your reputation**. Both are places where AI can quietly sit beside you and help you think more clearly, instead of trying to make decisions for you.

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

Use Case 1 of 2

Use Case 1: Researching a major purchase decision

On a Tuesday evening in late August, around 7:30 pm, Janet is sitting at her kitchen table with her laptop open and a legal pad next to her. She’s 58, lives in a mid-sized city, and her 12-year-old car has started making an alarming rattling noise. The mechanic gently suggested, “You may want to start thinking about replacing it.”

Janet has a decent savings cushion and is willing to spend money on something safe and reliable, but she hates car shopping. The last time she bought a car, she felt rushed and talked into features she didn’t need. This time, she’s promising herself: “I will not get steamrolled.” She’s looked at a few dealer sites, some review articles, and now her browser has 23 open tabs. Half of them contradict the other half.

She’s heard that “everyone” is using AI tools to research big purchases now - cars, solar panels, home renovations, even retirement software - but she doesn’t want a robot deciding for her. She wants the *kind* of help she used to get from a smart, trustworthy friend: someone to ask her good questions, organize her thinking, and lay out options without hype. She opens her AI assistant and, instead of asking “What car should I buy?”, she decides to treat it like a research aide, not a decision-maker.

Here is the exact prompt to use:

Copy and paste this into ChatGPT or Claude:

You are my personal buying advisor. I am considering a major purchase and I want a clear, practical overview, NOT hype.

Step 1 - Ask me questions: Before you give any recommendations, ask me up to 10 focused questions to understand my situation. Cover: - my budget range (including what “stretch” budget would be) - my location (city/region) - how I plan to use this [ITEM] - my priorities (e.g., reliability, safety, long-term cost, ease of use, resale value) - any constraints (e.g., no used items, must be available within 30 days, need financing, physical limitations)

After you ask your questions, WAIT for my answers. Do not give any advice until I’ve answered.

Step 2 - Once I answer, then do the following: 1. Summarize my needs in 5 bullet points, in plain language. 2. Based on public information up to your knowledge cutoff, outline 3 - 5 good options for me. For each option, include: - model/name and typical price range - key pros and cons for *my* situation - known reliability or quality issues (if any) 3. Explain what trade-offs I’m making if I choose cheaper vs mid-range vs premium options. 4. Give me a list of 8 - 10 specific questions I should ask a salesperson or contractor to avoid being taken advantage of. 5. Suggest a simple “next step” plan I can follow over the next 7 days to move toward a decision without feeling overwhelmed.

First, ask me your questions for Step 1. Then stop and wait for my replies.

Why this prompt works:

You’re telling the AI to be an **advisor, not a decider**, which matches how most people actually want to use these tools for shopping: to compare, narrow, and clarify, while you keep control of the final choice. You also force it to start with questions about *your* budget, priorities, and constraints, instead of jumping straight into generic product recommendations. That keeps the output grounded in your situation and less influenced by advertising-heavy sources.

One thing to watch out for

The AI doesn’t know your local inventory, current promotions, or very recent recalls. It’s working from patterns and public information, not live dealer stock or your neighbor’s experience. Treat what it gives you as a **starting shortlist and a checklist**, not a verdict. You should still cross-check prices and reviews on trusted sites and talk to at least one human - mechanic, contractor, or salesperson - to validate any big decision. Also be aware that its “knowledge cutoff” may mean it misses the newest models or rule changes; ask it explicitly if it might be out of date for your item.


Use Case 2 of 2

Use Case 2: Writing a professional decline email

It’s Wednesday morning, 9:15 am. David, 63, is semi-retired after a long career in healthcare operations. He still does occasional consulting for organizations he cares about. He opens his inbox and sees a message from Karen, a former colleague now leading a medium-sized nonprofit.

Her subject line is: “Could really use your help on this strategic review.” In the email, she explains they’re restructuring, there are politics involved, and she trusts him to give honest guidance. She asks if he can take on a 10 - 12 week project, starting “as soon as possible.”

David feels the familiar tug: he likes Karen, he believes in the mission, and ten weeks of consulting income wouldn’t hurt. But his wife had knee surgery last month, and he promised he’d keep his schedule light. He’s also quietly aware that these intense “special projects” tend to expand, chewing up evenings and weekends.

He wants to say **no** - or at least “not now” - without burning a bridge or sounding ungrateful. What he dreads is the socially clumsy email that either gives too much personal detail or sounds abrupt and transactional. He opens his AI assistant, but he’s wary of generic corporate language. He wants something that sounds like *him*, with the right blend of warmth and firmness.

Here is the exact prompt to use:

Copy and paste this into ChatGPT or Claude:

You are my professional writing assistant. I need to write a *polite but clear decline* email for a work-related request.

First, ask me 6 - 8 questions to understand: - who is asking (their role and our relationship) - what they are asking me to do (scope, timeline, level of commitment) - my real reasons for saying no (time, energy, conflict of interest, priorities, etc.) - how I would *like* this relationship to be in 6 - 12 months - any boundaries I want to reinforce (e.g., no rush projects, no weekend work, limited travel) - whether I’m open to a smaller or later engagement instead

After I answer your questions, write a draft email that: - opens with genuine appreciation (but not flattery) - briefly acknowledges the importance of their project or request - states my decline decision clearly in one sentence, without over-explaining - gives a simple, honest, and professional reason that fits my situation - reinforces the relationship (e.g., “I value our work together”) - if appropriate, offers a smaller alternative (e.g., a short call, a later start date, or a referral) - closes warmly but professionally, in my voice and at my level of formality

Use plain language suitable for someone in their 40s - 60s in a professional context. Avoid clichés like “Due to unforeseen circumstances” or “At this juncture.”

First, ask me your questions. Then stop and wait for my answers before drafting the email.

Why this prompt works:

You’re telling the AI to start by understanding **the relationship, the request, and your boundaries**, instead of jumping straight into a generic “template email.” That leads to wording that feels like **your** voice and situation - not something copied from a business writing book. By specifying the structure (appreciation, importance, clear decline sentence, brief reason, relationship reinforcement, possible alternative), you’re teaching the AI what good professional communication looks like and giving it guardrails to follow.

One thing to watch out for

AI-written decline emails can drift into either overly soft language that never quite says “no,” or stiff, legalistic phrasing that sounds nothing like you. Read the draft out loud and ask: “Would I actually say this?” If not, adjust it. Also be careful that you don’t let the AI invent details (“Given my upcoming travel…” when you’re not traveling) just to make the refusal sound smoother. Stick to reasons that are true, even if they’re simple, like “I’ve committed to keeping my workload light this fall.”


Know someone who spends too long on things AI could do in two minutes?

Forward Smarter by Thursday to three people who subscribe and I will send you my free AI Prompt Starter Pack: 20 ready-to-use prompts for everyday life.

Forward by email Open in browser Share on Facebook
Smarter by Thursday · By Dr. Rowan Hayes · drrowanhayes.com
Unsubscribe · View in browser

Don't miss what's next. Subscribe to Smarter by Thursday:
← Newer Issue 32: Smarter by Thursday Older → Issue 31: Smarter by Thursday
Bluesky
Powered by Buttondown, the easiest way to start and grow your newsletter.