Can Those ATM Receipts Laying on the Ground Tell Us Something About Money?

I have a money-related research project to run by you

Rocco Pendola
3 min readDec 6, 2021


Photo by Erol Ahmed on Unsplash

To see success writing on the internet, you’ve gotta be a little weird.

You can’t cut through if you just write about the same shit as everybody else using the same thoughts and ideas.

Occasionally, you feel like your weirdness hits a wall. As in, other people just aren’t having it.

I felt this way the other day after sending out a free post via my Never Retire newsletter, where I include thoughts not necessarily a fit for the Medium platform and expand on the core writing I do here.

Here’s what I wrote in that post —

When I was in college, I conducted a fair bit of research.

Because I didn’t have a budget for my research until graduate school, I had to put in serious work to develop something resembling a reliable sample for the questions I was trying to answer. So I stood on busy street corners and methodically used what researchers call a skip interval (stop every ‘nth person) to deliver my survey.

The rudimentary approach worked. I got two peer-reviewed academic publications out of it. You can see the abstracts here and here.

Anyhow, I never expected to be making this post, but I have had an idea for some money-related research for some time.

This might come off as weird, which is, in part, why I’m asking for your input.

I have always taken interest in the ATM receipts you see laying on the ground around banks, particularly ones with busy ATM machines.

I feel like they can tell you at least a little something about personal finance situations — at least in the aggregate. I wonder if you’d see significant variation in withdrawal amounts and balances on the basis of where the ATM is located, just as my previous research showed significant differences in sense of community and body mass index on the basis of neighborhood.

Anyhow, it’s probably obvious.

I want to come up with a plan to collect ATM receipts, aggregate them, and try to come to some loose conclusions based on the data…