Inspiration
Many founders test whether people are interested in an idea by collecting email addresses or measuring clicks. However, interest does not always mean that someone is willing to pay.
Worth A Try was inspired by this gap. We wanted to help founders validate not only demand, but also willingness to pay before investing significant time and money into building a product.
For the PayPal AI Hackathon, we extended Worth A Try with a price-testing feature powered by PayPal Sandbox. This allows founders to observe how users respond to different price points and whether they proceed toward a checkout experience.
What it does
Worth A Try helps founders validate early-stage product ideas through AI-generated landing pages and advertisements.
Users can:
- Enter a product or service idea.
- Generate a landing page and an advertisement with AI.
- Define a price range or manually specify price candidates.
- Show one price candidate to each visitor.
- Track landing-page visits and price-specific checkout progression.
- Compare results across different price points.
Each visitor is assigned one price from the configured candidates. The assigned price remains consistent for that visitor, allowing the founder to compare results more meaningfully.
The PayPal checkout experience uses PayPal Sandbox. No real money is charged. The purpose is to measure whether visitors proceed toward a payment flow at a particular price.
How we built it
We kept the original Worth A Try demand-validation workflow and added price testing as an integrated feature rather than creating a separate product.
The application uses:
- AI-generated landing pages and advertisements
- A price-candidate generator that divides a specified range into ten sections
- Automatic rounding to practical price points
- Manual price editing
- Visitor-specific price assignment
- Price-level visit and checkout-intent tracking
- PayPal Sandbox for simulated checkout flows
- A results dashboard for comparing performance by price
- Japanese and English user interfaces
The price assigned to a visitor is stored so that the same visitor does not see a different price every time they return. Owner previews show a representative price without affecting the experiment data.
Challenges we ran into
The main challenge was designing the experiment so that price was the primary variable.
If different visitors saw different landing-page copy or advertisements, it would be difficult to determine whether the price or the message caused the difference in behavior. We therefore kept the product description and advertising message consistent while varying only the price.
Another challenge was making the PayPal integration useful without charging real users. We used PayPal Sandbox so visitors can experience the checkout flow while ensuring that no real payment is made.
We also had to distinguish between actual experiment visitors and the founder previewing their own campaign. Preview activity must not contaminate the price-by-price results, so owner previews are excluded from the experiment metrics.
Accomplishments that we're proud of
We are proud that Worth A Try connects several steps of early product validation into one workflow:
- Idea generation
- AI-assisted landing-page creation
- Advertisement creation
- Demand measurement
- Price experimentation
- PayPal Sandbox checkout testing
- Price-specific result comparison
The most important accomplishment is that PayPal is not merely used as a payment option for Worth A Try. It is used as part of the validation methodology: the checkout progression becomes a signal of how strongly a visitor may value the proposed product at a given price.
What we learned
We learned that collecting registrations and measuring clicks are useful, but they do not tell the whole story. A visitor may be interested in an idea but still be unwilling to pay for it.
We also learned that price experiments require careful control of other variables. Keeping the product message consistent is essential when comparing different prices.
Finally, we learned that a sandbox payment flow can be valuable even when no actual transaction takes place. It allows founders to test a deeper level of commitment than a simple click or email registration.
What's next for Worth a Try
Next, we plan to improve the experiment analytics and make the results more statistically useful.
Potential improvements include:
- More detailed conversion funnels
- Confidence intervals and sample-size guidance
- Better comparison of price segments
- Support for multiple currencies
- Improved campaign and visitor management
- More advanced AI-generated experiment recommendations
- A mode that measures checkout-button clicks without opening the PayPal checkout page
- Support for real payments only after a product is ready to launch
We also envision Worth A Try being used in entrepreneurship education and public startup-support programs. Municipalities, universities, and startup accelerators could use it to help students and aspiring founders test their ideas with real-world evidence rather than relying only on intuition.
By incorporating Worth A Try into university entrepreneurship courses or local government startup-support initiatives, participants could learn how to formulate hypotheses, design simple experiments, interpret customer behavior, and make evidence-based decisions about their businesses.
Our long-term goal is to help founders move from “People say this is interesting” to evidence-based decisions about demand, pricing, and product development.
Built With
- cloudflare-d1
- cloudflare-workers
- css
- drizzle-orm
- html
- javascript
- openai-api
- paypal
- sandbox