I suggest reading this discussion. Has some really nice answers.

You always spend time testing. The beginner approach is purely manual testing.

Writing tests certainly takes time. It also saves time from less manual testing initially and faster finding of issues and preventing of regressions later.

I would highly recommend you to watch this, not about testing itself but the costs and tradeoffs regarding to testing:

https://www.youtube.com/watch?v=uQUxJObxTUs

I believe in Continuous Testing (no that's not a thing. Yet…)

I write unittests. A lot of them. If I don't have the time, I still write tests for the happy path, so I can be sure it won't go wrong. If I work with code that has no tests, I add some, at least for the feature/bugfix I work on.

Then I tend to add functional tests. They are more scarce, as it takes a huge amount of time to implement one, and I'm a software developer, not a tester. That's a separate job.

If other software interacts with mine, e.g. through an API, I also do integration tests.

But most importantly, I test in production. Now I can hear a lot of you screaming, but Testing in Production (TiP) is a thing. We have select test cases that are safe to run against the production site, and we do several times a day. Also, we are in a lucky position to not just develop, but also use our software on a daily basis. Thanks to this, we are among the first few to discover anomalies.