It's about economics not accounting.
Once you get into cost comparisons you've already lost. AWS is purpose built for an accounting narrative.
To illustrate what a dead end accounting is, accounting can't explain things like, "I spend money today to possibly make money more than 1 year in the future." That's basically every startup and accounting doesn't have a story for it.
Of course a free tier looks good accounting-wise. How does one "account" for lock-in though? If you can't figure that out you will not convince bean counters to consider alternatives.
And good luck teaching economics (as opposed to accounting) to bean counters.
Here are a few compelling explanations: the cheapest bid (i.e. free tier) is always the worst one. Some people are so addicted to accounting storytelling - as a way to organize their world, a whole philosophy - they are actually always convinced the cheapest bid is always the best one.
Another: AWS is so overpriced, the $50k you spend on developing against AWS services to use $500k of "free credits" will deliver less value than $50k you spend developing for a single vertically scaled beige box computer with simple ergonomics and running it.