Smarter by Thursday logo

Smarter by Thursday

Archives
Log in
Subscribe
August 8, 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

Last week, I kept thinking about two places where AI is genuinely useful for normal people: **researching a major purchase** and **writing a professional decline email**. Those sound like small tasks until you are the one staring at a too-expensive decision or trying to say “no” without sounding cold. For people in their 40s, 50s, 60s, and beyond, the value of AI is not magic. It is having a steadier second brain when the stakes are real and your time is limited.

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 Thursday morning at 9:10, Linda, 58, is sitting at her kitchen table with three browser tabs open and a headache she cannot ignore. Her old refrigerator started making a grinding noise last week, and now the repair estimate is almost high enough to justify replacement. She does not want the cheapest model. She wants the one that will still be running in ten years, with the least chance of expensive regret. But every website says something different, and every review sounds either suspiciously enthusiastic or suspiciously angry. What Linda really needs is not more noise. She needs a way to compare options, identify the tradeoffs, and focus on the few facts that matter for *her* situation: reliability, energy use, repair history, size, and total cost. That is where a well-written prompt helps.

Here is the exact prompt to use:

Copy and paste this into ChatGPT or Claude:

```text I need help researching a major purchase before I spend real money.

I am deciding whether to replace my refrigerator, and I want a practical comparison that helps me avoid buyer’s remorse. Please act like a careful research assistant, not a salesperson.

First, ask me up to 7 questions that would change the recommendation, such as budget, kitchen size, preferred features, how long I plan to stay in this home, and whether I care more about reliability, energy savings, or convenience.

After I answer, do the following: 1. Summarize my priorities in plain English. 2. Give me a list of the most important factors to compare for this purchase. 3. Create a comparison table for 3 to 5 refrigerator options or categories that fit my needs. 4. Explain the tradeoffs in simple language, including what I would gain and what I might give up with each option. 5. Point out common sales tricks, hidden costs, and questions I should ask before buying. 6. Give me a short final recommendation based on my priorities, not a generic “best choice.”

Keep the tone direct and practical. If you are uncertain about any detail, say so instead of guessing. ```

Why this prompt works: It works because it forces AI to slow down and gather context before jumping to conclusions. It also tells the model exactly what “good” looks like: a comparison, tradeoffs, hidden costs, and a recommendation tied to *your* priorities. The strongest part is that it asks for uncertainty to be stated plainly, which is essential when you are comparing expensive products and do not want polished nonsense.

One thing to watch out for AI can sound confident even when the underlying information is outdated or incomplete. That matters with products, warranties, and model changes. For anything expensive, use the prompt to narrow the field, then verify the final details with current retailer specs, manufacturer information, and a real human if needed.

Use Case 2 of 2

Use Case 2: writing a professional decline email

On Tuesday at 4:40 p.m., Marcus, 63, is staring at an email from a former colleague who wants him to join a committee, volunteer for a project, and “help shape the future” of something he has no bandwidth for. He likes the colleague. He respects the organization. But he also knows that saying yes would mean one more obligation layered on top of work, family, and the hours he has already promised elsewhere. Marcus does not want to write a reply that sounds rude, slippery, or overly dramatic. He wants a message that is professional, kind, and final enough to stop the follow-up emails. That is exactly the kind of situation where AI can help you draft a clean refusal without making it sound robotic.

Here is the exact prompt to use:

Copy and paste this into ChatGPT or Claude:

```text Draft a professional decline email for me.

I need to say no to [invitation/request] from [person or organization]. My goal is to be polite, brief, and clear. I do not want to over-explain or leave room for pressure.

Use this context: - Relationship: [colleague, client, manager, friend, vendor, etc.] - Reason for declining: [too busy, not the right fit, already committed, not enough expertise, etc.] - Tone: warm, respectful, firm - Length: short, about 80 to 130 words - Style preference: [direct / friendly / formal / casual-professional] - Extra note: [I may want to leave the door open for future opportunities / I want this to sound final]

Write 3 versions: 1. A very short version. 2. A warmer version. 3. A firmer version.

Then tell me which version is safest if I want to avoid sounding rude. ```

Why this prompt works: This prompt works because it gives the model the pieces that actually change the email: who the message is for, how firm it needs to be, and whether you want to keep the relationship open. Asking for three versions is especially useful because “professional” is not one thing. A decline to a client is not the same as a decline to a neighbor, and the prompt helps AI adjust to that difference.

One thing to watch out for AI-written refusal emails can sometimes sound polished but oddly empty, like they were written by a committee that never had to be the one saying no. That can be a problem if the message needs real warmth or if the relationship is sensitive. Read the draft out loud before sending it. If it sounds like a politician dodging a question, tighten it until it sounds like *you*.

Know someone who spends too long on things AI could do in two minutes? Forward this to three people who subscribe and I will send you my free AI Prompt Starter Pack: 20 ready-to-use prompts for everyday life.

Next Thursday how to use AI to clean up messy notes and turn them into something you can actually use.

By Dr. Rowan Hayes


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 32: Smarter by Thursday
Bluesky
Powered by Buttondown, the easiest way to start and grow your newsletter.