"optimize for promotion" is one way to frame but, but alternatively I'd suggest to only work on things you can prove to have value. You can do this by either ac

"optimize for promotion" is one way to frame but, but alternatively I'd suggest to only work on things you can prove to have value. You can do this by either actively convincing others it has value (politics) or more passively always relying on external validation of your judgement. For example, if you see a problem, you should socialize a solution to that problem. If the feedback you get is "Yes, this is a problem, and that solution would be valuable", then go for it. Of the feedback is more meh in nature, then either convince people it is valuable or move on to something else. Be organized in this process, write RFCs and design docs, document your conclusions. Then when it comes time to discuss impact, you will know (and can demonstrate) that you worked on impactful things.