I came to the same conclusion after running small-changes A/B test. But I think it still makes sense for smaller projects to make "big" A/B test, like testing

I came to the same conclusion after running small-changes A/B test. But I think it still makes sense for smaller projects to make "big" A/B test, like testing a completely different homepage/landing page with different copy and marketing approach. Another example of useful A/B test on small scale: for registered users, testing if an automated "support" email is helping moving customers to the next step of your funnel (A: send email B: don't send email. Measure the conversion of that specific funnel step)