Smarter by Thursday logo

Smarter by Thursday

Archives
April 26, 2026

Issue 17: Smarter by Thursday

Smarter by Thursday — Issue 17

Issue 17 · week of April 20, 2026

Smarter by Thursday

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

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

There are two things you're probably doing wrong this week, and both of them cost you time and money. The first is how you're planning trips - you're either ignoring AI entirely or trusting it blindly, which leads to itineraries that look perfect on screen but fall apart in real life. The second is how you write complaint letters. You're being too polite, too vague, and too willing to accept the first response you get. Both of these are fixable, and both matter more than you think. A good road trip plan saves you from exhaustion and wasted gas. A well-written complaint letter gets you actual results instead of a form letter. Let's fix both.

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

Use Case 1 of 2

**Use Case 1: Planning a Road Trip Using AI**

It was a crisp Tuesday morning in early April when Marcus Chen, a 62-year-old retired accountant from Portland, decided to finally take that cross-country road trip he'd been talking about for three years. He had two weeks in May, a sensible budget, and a vague idea of driving from Portland to Santa Fe. He opened ChatGPT and asked it to "build me a perfect two-week road trip itinerary from Portland to Santa Fe, hitting the best spots." Within thirty seconds, he had a detailed fourteen-day plan with attractions, restaurants, and hotels. It looked beautiful. It looked efficient. It looked like it would kill him. The itinerary had him driving eight hours one day, then hitting five museums the next morning, then a three-hour drive to a hiking trail at 2 p.m. There were restaurants he'd never heard of, some of which didn't actually exist. He had no buffer time for traffic, no downtime for his aging knees, and no backup plan if something went wrong. Marcus almost gave up on the trip entirely. Instead, he learned to use AI differently.

Here is the exact prompt to use:

Copy and paste this into ChatGPT or Claude:

"I'm planning a road trip from [START POINT] to [END POINT] over [NUMBER] days. I'm interested in [TYPES OF ACTIVITIES: museums, hiking, food, history, etc.]. I drive slowly and like to stop frequently. I get tired by 4 p.m. and need downtime. I want to avoid big cities / I love cities [choose one]. Build me a day-by-day itinerary that clusters activities by location so I'm not zigzagging. Include driving times between stops, and add a 30-minute buffer between each activity. Then, stress-test this itinerary: tell me which days are too packed, which transitions are too tight, and what could go wrong (bad weather, road closure, attraction closed). Finally, give me three backup plan options for each day in case I need to change things."

Why this prompt works: You're giving AI specific constraints about your energy level and preferences, not just asking for a generic itinerary. You're asking it to cluster activities geographically, which saves gas and sanity. You're explicitly requesting buffer time and backup plans, which AI won't do unless you ask. Most importantly, you're asking it to *critique its own work*, which forces it to catch the problems before you hit the road.

One thing to watch out for AI will still occasionally place attractions in the wrong neighborhood or give you driving times that don't account for real traffic conditions. Always verify addresses on Google Maps, check current hours of operation for museums and restaurants, and add 20 percent to any driving time estimate AI gives you. A 2025 study found that 90 percent of AI-generated itineraries have at least one significant error. That's not because AI is useless; it's because AI doesn't know that your local highway has construction, or that the "charming local restaurant" closed six months ago.

Use Case 2 of 2

**Use Case 2: Writing a Complaint Letter That Actually Gets Results**

It was Thursday afternoon when Patricia Alvarez, a 58-year-old small business owner, received her credit card statement. She'd been charged $847 for a hotel room she'd cancelled two weeks before the stay. The hotel's cancellation policy said "free cancellation up to 48 hours before arrival," and she'd cancelled with four days' notice. She called the hotel's customer service line, got transferred three times, and was told there was nothing they could do - the charge had already been processed. Frustrated, she sat down and wrote an email to the hotel manager. It was polite. It was reasonable. It explained the situation clearly. It got no response. A week later, she tried again with a different tone: direct, specific, and clear about what she wanted and what the consequences would be if she didn't get it. That email got a response within four hours, an apology, and a full refund.

Here is the exact prompt to use:

Copy and paste this into ChatGPT or Claude:

"I need to write a complaint letter about [SPECIFIC PROBLEM]. Here's what happened: [CLEAR TIMELINE OF EVENTS]. Here's what I was promised: [WHAT THE COMPANY/PERSON SAID THEY WOULD DO]. Here's what actually happened: [WHAT WENT WRONG]. Here's the impact on me: [SPECIFIC CONSEQUENCES - money lost, time wasted, damage, etc.]. I want [SPECIFIC RESOLUTION - refund, replacement, repair, etc.]. Write me a professional but firm letter that: 1) States the facts without emotion, 2) Includes specific dates and amounts, 3) References the original agreement or policy, 4) Explains why their response is unfair, 5) Clearly states what I want, 6) Includes a reasonable deadline for their response, 7) Mentions that I'm prepared to escalate this [to credit card company / small claims court / public review sites / regulatory agency - pick what's actually true for your situation]."

Why this prompt works: You're not asking AI to be nice or to apologize for bothering them. You're giving it the facts and asking it to build a logical case. The key is specificity: dates, amounts, names of people you spoke with, and references to the actual policy or agreement. You're also being clear about what "winning" looks like and what happens if they ignore you. Companies respond to this because it signals you're serious and you know your rights. Vague complaints get vague responses. Specific complaints with clear consequences get action.

One thing to watch out for Don't threaten anything you're not actually prepared to do. If you say you'll take it to small claims court, you need to mean it. If you say you'll dispute it with your credit card company, be ready to actually do that. Companies can tell when you're bluffing, and it undermines your entire letter. Also, don't use this prompt to be cruel or unprofessional - the goal is to get results, not to make someone feel bad. A letter that's firm and factual will always work better than one that's angry.

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