[mikekoscinski]

It looks that way to me too.

I converted the conversion rates from unique-trial-paid for each scenario. It looks like 0.846% for credit card required upfront and 0.264% for no credit card required. So *net conversion is ~3.2x higher when a credit card __is__ required upfront.*

Showing my work:

I eyeballed each bar chart and turned it into a data table. I'm getting median unique-to-trial conversion rates of 3.1% (credit card required) and 6.1% (credit card not required). I'm then adjusting these to account for the % of "don't know" responses. So, adjusted medians of 1.8% (cc required) and 3.3% (cc not required).

Following the same process for trial-to-paid yields medians of 47% (cc required) and 8% (cc not required).

Multiplying each pair yields 0.846% (cc required) and 0.264% (cc not required) unique-to-trial-to-paid.