[aazaa]

> When I finally did user interviews, I uncovered that an unusually high amount of people have “travel funds”: separate bank accounts they use for travel. I ran a statistically significant survey for the United States. I found that more than 25 million adults in the U.S. use a separate bank account for travel, but almost no one I talked to had automated and optimized their process for transferring money into and out of their travel fund.

That's surprising. I would have suspected that more than 75% pay for their travel on a credit card. They decide to travel first, then pay for whatever piles up later. Or maybe they just let the balance accrue.

How did you conduct a "statistically significant survey" on this for the US? Maybe there's a startup idea there.

[mpc75]

I used Google Surveys which handles the cleaning of responses for representative sampling for U.S. 1,500 people took the survey but only 1,128 were recorded after cleaning.

[staticautomatic]

Typically this just means the naive approach: 385 or so people assumed to be randomly sampled, which yields results at 95% confidence with a 5% margin of error.