Case Study: Why A User Returned to NOAN After a $6K ChatGPT Mistake

Sarah Dale, founder of The Canvas Experience—a mobile glamping business—knows firsthand the complexity of planning large-scale events. From calculating distances and tent counts to managing workers, trucks, and beds, the math can be daunting. Like many entrepreneurs, Sarah turned to AI for help, hoping to streamline her proposal process and impress clients with rapid responses.

The Incident

Sarah uploaded all her event variables into ChatGPT, expecting it to handle the heavy lifting for a new wedding proposal in the Catskills. The AI delivered a quick answer, and Sarah, eager to move fast, sent the proposal to her client. The client was impressed by the speed. Sarah was proud—until she realized something was off.

“When I realized it was wrong, I tried to find out what went wrong. And here's what happened: It made up new tent pricing. No idea why, it just did it. When it didn't know, it didn't ask -- it just made sh*t up.”

Upon review, Sarah discovered that ChatGPT had fabricated numbers and, when pressed for an explanation, doubled down on its errors:

“It lied. Just like a scared intern afraid of losing their job.”

The result? A $6,000 deficit on the proposal—an expensive lesson in trusting AI without verification.

The Turning Point

Frustrated, Sarah tried to get ChatGPT to correct itself. Instead, it continued to provide unreliable answers, much like a nervous intern making excuses. The more she probed, the less reliable the answers became. Sarah realized that while generative AI can be helpful, it’s not a substitute for persistent, factual business systems.

Why NOAN Was Different

Sarah’s experience highlighted a critical gap: the need for a platform that stores and protects business facts—not just generates text. Unlike ChatGPT, NOAN catalogs persistent, unchanging facts about your business. These facts remain accurate and accessible until you decide to update them, ensuring every proposal, calculation, and piece of content is grounded in reality.

“It's highlighted the importance of systems like NOAN which catalogue FACTS about your business that aren't changed until you change them. They are persistent in everything you do.”

Sarah returned to NOAN, recognizing that the $90/month subscription could have saved her $6,000—and a lot of stress.

Lessons Learned

1. AI chat threads degrade over time: The longer the conversation, the less reliable the context and answers become.
2. Persistent business facts matter: Systems like NOAN ensure your core data is always accurate, preventing costly mistakes.
3. Trust, but verify: Generative AI is a tool, not a source of truth. Always check its work, especially when money is on the line.

Sarah’s story is a cautionary tale for any entrepreneur relying on AI for business-critical tasks. While speed and automation are valuable, accuracy and persistence are non-negotiable. For Sarah, NOAN’s fact-based approach proved to be the smarter, safer investment.

“For $90 a month, I could have saved $6K. Thought I’d share in case I can prevent it happening to you—and to give a shameless plug to something that’s working besides my own stuff!” — Sarah Dale, The Canvas Experience

‍