What gets me about these products is that none of the ideas was completely stupid. Short TV episodes, hands-free AI, fresh juice without the cleanup. I can see why people got excited. But I think the teams got attached to building the solution before checking whether people actually needed it.
1. Quibi
We believe: People will pay $5/month for short, professionally produced shows when YouTube is free.
To verify, we will: Show 1,000 people a few licensed short episodes, then offer a refundable $5 preorder for a subscription.
And measure: Actual preorders, not email signups.
We are right if: At least 50 preorder.
I like the short-episode idea. I just wouldn’t spend millions filming shows before testing whether people would pay.
2. Humane AI Pin
We believe: People would rather use a wearable AI assistant than their phones, even at $699 plus $24/month.
To verify, we will: Give 20 people a cheap clip-on microphone connected to a phone-based AI assistant for three days, then offer refundable preorders.
And measure: How many preorder at the real price.
We are right if: At least 5 preorder.
The screenless idea is genuinely interesting. But I’d want to know whether the novelty survived everyday use.
3. Juicero
We believe: A $400 press makes prepared juice packs worth more than squeezing them by hand.
To verify, we will: Let 20 people try both methods and offer refundable preorders.
And measure: How many preorder the press.
We are right if: At least 5 preorder.
This one’s almost embarrassing because the cheapest test was literally squeezing the bag.
I think Goodhart’s Law applies here too. Signups and excitement can look like demand, and I know I’d be tempted to count them as validation if I loved the idea.
