I'd say having kids is also really good and worthwhile if you raise them into responsible, productive adults.
Why would influence have to always increase? Kids have a lifespan as well, it's just much longer than most projects.
The piece of software I'm currently working on will almost certainly not be doing squat 20 years from now. It will have been replaced with something else or upgraded beyond all recognition. And if I wasn't there to write said code some other software developer would be hired to do it. My personal contributions will have zero or asymptotically approaching zero influence 20 years from now. Kids on the other hand...
Kids aren't entirely selected at random, you select your partner based on impulses evolutionarily designed to produce stronger offspring. Then you spend over a decade shaping their behavior, whether you choose to do so directly or not. If you're a responsible parent you teach your kids what you know so that they can build on it as they see fit, when they grow up. And then they pass along a diluted form of said lessons to grandchildren. Now there's no guarantee of this happening, tragedy can strike and you can mess it up. But it's an effective enough mechanism that, for most people, your indirect influence through raising kids will be far greater over time than your direct influence via your work. It's one of the ways generational wealth is formed.
If you can take an honest look at your work and say:
1. If I don't do this no one else will, or they'll do it substantially worse
and
2. Its direct or indirect impact will still be felt 70 years from now
Then sure, don't have kids. And I'm not being facetious, if you're working on some unique device or policy that shows real promise at saving/fundamentally improving lives and your leaving would set development back years or worse, that qualifies. Most people aren't doing stuff like that though, in part because simply being in a position to do so is largely luck. In that case, kids are probably a better bet.