You could argue that self-hosting matters now more than ever in an age of creeping data permissions and rampant monetization. Over the past year, self-hosting services have taught me how to claw back some independence from Big Tech rather than feeling helpless.

However, after trying out many different apps, there are some that just aren't worth self-hosting for me. These factors help me decide whether to avoid a self-hosted app entirely or whether to cut it loose from my setup...

Seeing how much value it adds to my daily routine

It's subjective

There are tons of self-hosted apps you can try out, but only you can really decide how much value they really add to your life. One of the benefits of self-hosting is that it can free you from subscription fees, but sometimes an app is worth paying for, and you might not find that same value in a self-hosted version. It's context-dependent.

For me, self-hosting simpler apps has provided the most value, as well as certain backup and digitization solutions. Other times, I end up removing a tool from my setup because I don't find it all that useful.

The value a service provides is subjective, so you often can only evaluate it by trying it out yourself. I definitely recommend trying out services to see how they fit into your own routine.

Sometimes it's worth jumping through compatibility hoops and frequent troubleshooting for an app you really love. But that effort is wasted when it comes to something that doesn't provide much joy or use.

How detrimental downtime would be

I skip self-hosting critical services

While your IT knowledge will factor into what types of services you can self-host, there is a reason that even the most knowledgeable people don't self-host everything. For example, plenty of people don't bother with self-hosting email due to the detrimental effect of downtime.

When I decide whether I want to self-host a service, I factor in what the potential effect of downtime would be. For something like Immich or Jotty Page, the services going down are mostly just inconvenient for me. I have more than one copy of essential files, and if I can't access the service, I have a backup system in place.

In contrast, if I were to self-host my own website, downtime would affect traffic, which in turn affects advertising prospects. I've never considered self-hosting email because not receiving certain emails would mean I lose out on important information and potential opportunities. While self-hosting a calendar server is totally possible, I worry too much about missing appointments and meetings.

Whether I spend more time troubleshooting it than using it

No sunk cost fallacy for me

There are times that I set up a self-hosted service, enjoy the experience, but then keep running into issues that impact the core functionality. No matter how much I like the premise of a service, there's no point in using it if there are constantly bugs that break its core features.

I recently had this experience with a self-hosted container management platform. The UI was modern, and it had some really cool features. But when I spent more time using its different parts, I discovered one issue after another. Fixing one issue would sometimes result in another feature breaking.

Based on the GitHub forums and Discord server, this wasn't down to my own error, but issues with the way the platform was built. I still hope to see it grow into something great one day, but for me personally, it's a bit too unstable currently to be worth my time.

When you self-host, some troubleshooting is to be expected. I've set up services and gotten features working, only to have them fail a few days later. We'll all have our own tolerance threshold based on things like how much spare time we have. Once I've spent multiple days troubleshooting a service only for continued issues to pop up, I usually take a break and decide whether it's worth revisiting later.

Checking if there are any security risks

Some functions are more sensitive than others

When it comes to self-hosting, it's also important to know your limitations. While I know not to insecurely expose my services to the internet, I don't know enough about cybersecurity to host services with sensitive functions.

So when it comes to self-hosted services that could risk my privacy and safety through misconfiguration, I steer clear of them. Even the services that are only accessible on my home network don't have sensitive documents or information on them, just in case there's a vulnerability I've missed.

You can always learn the skills you need to host certain services, but for me, it's just not worth the risk to try to do this in my case. So when it comes to self-hosted security software or even something like a self-hosted password manager, I just don't have the appetite for the uncertainties.

Decide your dealbreakers when it comes to self-hosting

The things that make me avoid a self-hosted service may be different for others. Some people prefer to control as much of their data as possible and have the skills to make that happen. For others, convenience may be the most important factor.

I would say that while you should feel free to experiment with different services, it's worth knowing your limitations. Also, don't feel pressured to keep using a service just because you've sunk a lot of time into it. Stick with what works for you and brings value to you.