I personally do value velocity not in number of lines, but in terms of unblocking users and other team members. So: 1) user wants X done. Focus on that. Help

I personally do value velocity not in number of lines, but in terms of unblocking users and other team members. So: 1) user wants X done. Focus on that. Help the user X done. First iteration you have something crappy, doesn’t look great, too simple. That’s okay. Did it get X done. Cool. If you shipped some pretty UI but users don’t actually use it since it doesn’t solve the problem. That’s not productivity. 2) now that there are users using the tool, watch them use it. Ask questions, look at th me dashboard. Optimize their flow to do X. Make the UI delightful. We know with good certainty that this is a decent solution. Make it fast. Add tests. Harden it up so someone else doesn’t break it accidentally. So if someone added two lines of using some existing node module that solves the user’s problem, those are very productive two lines.