Crypto Card Research Notes

I run a few public research notes on crypto cards in addition to Sweepbase . People sometimes ask why I do not just put everything on the main site. The honest answer is that the side notes are where I test angles before they earn their way into the catalog. The main site has to stay legible. Every page is rated, scored, and sortable, which means I cannot publish an angle there until I have run the math across all 139 cards. The notes are where I write up a single observation, link to the source data, and see if anyone pushes back. What lives where The longer essays go on Substack. The most recent one explained why most cashback claims do not survive a careful read of the tier requirements: Crypto card cashback is mostly fake . The technical writeup of why I built the data layer on a CSV file rather than a database is on Dev.to: My Next.js 15 aggregator runs on a CSV file instead of a database . The shorter notes go on Telegraph. They are anonymous and that lets me write more d...

Last week I posted a note on Telegraph about why cashback rates on crypto cards are usually misleading. After publishing it I had a few people ask me to put hard numbers behind the abstract claims. So this is a short follow-up rather than a fresh argument. The original post is here: When crypto card cashback rates are misleading . The idea was that headline rates rarely match what hits your account because of tier requirements, monthly caps, category exclusions, and token volatility. What I want to add is one example. Take a card that advertises 5 percent cashback. The fine print usually says something like: 5 percent on grocery and gas categories, 1 percent on everything else, capped at $50 per month, requires staking $1,000 in the issuer's token. Here is what that looks like for someone spending $1,500 a month with $400 in grocery and gas: 5 percent on $400 = $20 (under the cap, so the full $20 counts). 1 percent on $1,100 = $11. Total: $31 in cashback. Effective rate = $31 ...