The worst expense is a project which you won't admit you'll never get back to but you're still paying that $50 a month for the collective expenses.
When creating new projects (especially complex ones with multiple services involved) I always try to think about the long-term maintenance and scalability aspects, so I replied:
Excatly, I was gonna say 😀 You must think in long term in the very beginning when choosing the stack and weight all the ifs for the possible abandonment due to new projects with higher potential, etc 😉
For example, I'm very happy with the stack I chose for my most recent project devQuizzes (a quizz platform where users can test their dev knowledge in over 500 questions across 20 categories):
I picked GitHub to store all the data (each question is a seperate .md file in the repo), Airtable for users to submit new questions and submit edits for the question (I use their API for this, so the inputs are stored in their Sheets), Vercel to host it all (which I absolutely love, no promo btw) and for the domain I created a subdomain quizzes.madza.dev for the only domain I own madza.dev (I use NameCheap, which I found in this discussion, thanks to you guys), meaning I can still buy a separate domain and set up redirect to it if the project starts to become something serious.
This means that the total maintenance costs for the project is zero and I am not limited to move to other projects if I want to in the future without sacrificing my bank account.
The question for this discussion is:
Do you have any abandoned (inactive) projects that you must pay for monthly/yearly to keep them running? Like hosting, domain, database, storage and other costs.
Also, what are the key lessons you learned from it (what would you do different when setting up the same project today)?