Did you take your consent for sending marketing emails? Probably not. Something to keep in mind. Most will tolerate it, if you are providing value.

That being said, send regular text-emails and not market-y emails from the system you have in place. Query all email addresses and sending email in a loop ain't too much work. If you have a lot of customers, do that in batches.

Stick with what you know when you want reliability. Want to build products fast? Use the tech stack you are familiar with. Want to improve your tech? Learn new things. Choosing one over the other is very scenario specific. If you use new tech for building something fast, you would be inefficient with time but you will learn. It's a trade-off really.

I kinda hate JavaScript, but I love Vue.js as to how easy it has improved my skill set to do frontend code. So, if I would have stuck with JavaScript, I would have done myself a disservice in the long run :)

Here's what I do:

Assuming I am starting from "Unproductive" state, no tool/system has been the right fit.
I start with paper and pencil to note down tasks & just start with something. Momentum helps.
And once I have momentum, I will observe what I am attempting to do, where I can use the tech in my productivity system. Then I will just go get that. This could be a custom thing I put together or whatever. These days, I use whimsical (drawing tool) to manage some of my mindmaps and todo systems. No kidding. You will bound to grow out of it, and just adapt as you go. And when I fall off the wagon after weeks/months, I reset and start with pen and paper again.

Moral of the story - Don't find a tool to fit into it. Deeply observe how you work and find tech bits that help you in your evolving process.

Two questions: How much downtime can you afford? What's your budget to fight against that downtime? Essentially a question to figure on the basis of how much loss does downtime incur you.

Basic idea would be to replicate the entire stack over AWS, test it and once that's functional, you can do something like pausing the live site to only be read-only mode (maintenance) and then update the database(s) over at AWS by taking a fresh dump from GC.
Make changes in application to make GC environment connect to AWS dbs and use them.
Then make the site go out of maintenance and all db updates are happening over at GC. Simply update the DNS at this point to move the traffic over to AWS. You can lower your TTL for DNS a few days before to make this go fast. That's pretty much it but definitely not a single solution for all needs, most should fit.

Are you using Stripe Connect or just Stripe? Former is basically build for marketplaces where you charge from your account and transfer Sellers' share to their stripe account after deducting your fees.

If they are able to grab customers off that post, pay attention to what customer need they are serving which is covering customers? Learn from them & don't worry about competitors. The pie is big enough and you just need to grab a slice for yourself.

Does your neck not hurt with that big of a vertical screen? When I rotate my 27" screen to be vertical, its tall enough to cause my neck to hurt. Yours look even taller.

Standing desk: $4K Rig, $100 Tables XD

I haven't heard about Render before. Looks cool, how did you setup your infra on it?

Hey! At some point you are bound to hit a ceiling where you are trying to use something which wasn't the original use-case, so you either move to another system or find a workaround.

Have you talked to their support and explained your use-case? May be its a soft limit that they can raise for you? If not, then consider how hard it would be for you to use multiple Vercel accounts abstracted away under your implementation? Kinda hacky, but that's what hackers do.