Given my familiarity with Vercel, I initially wanted to deploy my Hugo blog there.
But the process was just too difficult. Following the official guide led to a webpage that displayed some, but not all, features of my theme.
I tried a few troubleshooting tips from various sources, but without a deep understanding of how Vercel works, couldn’t find a fix.
Online discussions seem to put the finger of blame on which Hugo version is used by Vercel: but without knowing the full context it was hard to get a feel for whether that is right.
I then changed tracks to starting from a Hugo starter provided by Vercel itself. It worked perfectly well with the default theme, but once agin, the moment I switched to other themes, problems would surface.
It only occurred to me 1-2 hours in that there is another way of doing things.
I chose Vercel in the first place because I thought I could save time with my familiarity with Vercel in general.
But given it is a hobby project I don’t have any solid reason to choose Vercel over anything else. The time I had spent on finding various fixes for the Vercel deployment already outweighs any time savings from familiarity. Time to give up the struggle and start afresh.
Hugo lists a wide range of options for deployment, each with its own dedicated tutorial. After trying Gitlab and finding the set-up too complex, I settled on Netlify and the process went very smoothy from there.
Overall, in this situation, I think I struggled slightly too much. The desire to stick to a well-known deployment strategy is understandable.
But there must be a point where the costs outweighs the benefits: especially if (as in my case) main goal is just to get a website up-and-running, as opposed to learning anything very deep about the process of deployment as such.