🧠 Generating Evidence-Based Ideas

Read the FULL prompt directly on ChatGPT

My learnings from prompt experiments:

👉🏾 Randomise constraints: Ask ChatGPT to randomly select behavioural principles (e.g., MINDSPACE) and then challenge it to generate ideas within those constraints. Giving it limitations paradoxically leads to more creative solutions.

👉🏾 Stacking frameworks: Use multiple frameworks sequentially to force different perspectives (e.g. 6 Thinking Hats). This keeps ChatGPT from getting stuck in predictable patterns and pushes it towards fresh insights.

👉🏾 Reflective thinking: Instead of letting ChatGPT run wild, insert pause points where you can ask it to consider ‘unintended consequences’. This encourages the model to develop sophisticated risk-mitigation strategies.

Sample ChatGPT Input:

We are applying behavioural science principles as constraints to think more creatively.

First, acknowledge that constraints enhance creativity and innovation.

Then, ask me to provide the MINDSPACE framework so we can use them as pathways for behaviour change.

Sample ChatGPT Output:

🔍 Challenge:
How might we make it easy for credit card users to automate full-balance payments without requiring manual setup?

🧠 Behavioural Principles Used:

Norms – Leveraging social proof to show that most responsible users automate full-balance payments.

Defaults – Pre-selecting full-balance autopay to make it the easiest option.

💡 Key Insight:

Most people stick with default settings and are influenced by social norms. By making automatic full balance the default choice and reinforcing it with peer behaviour insights, we can drive adoption while maintaining user trust.

🚀 Create Solution:

New credit card customers are softly defaulted into full-balance autopay, with messaging highlighting that most financially savvy customers use this highly recommended feature by the bank to support their financial wellbeing. Users can confirm to accept the savvy option or opt out with a single tap.

🎯 Benefits:

✅ Effortless setup – Eliminates manual configuration, reducing friction.
✅ Social reinforcement – Encourages users to follow responsible financial habits.
✅ Avoids unnecessary fees – Helps users avoid interest charges through automation.
✅ User control – Maintains transparency and allows easy opt-out.
✅ Adaptive safeguards – Prevents overdrafts by adjusting payment levels dynamically.

⚠️ Risk and Mitigation:

Risk: Customers' perceived lack of control 
→ Mitigation: Use a soft default instead of a hard auto-enrollment, requiring one-tap confirmation.

Risk: Overdraft concerns 
→ Mitigation: Implement an adaptive payment system that adjusts based on balance levels.

Risk: Regulatory pushback 
→ Mitigation: Keep clear documentation on how this change supports financial wellbeing and highlight easy opt-out options for users.