You can't exactly decide what to reward. A lot of the important aspects of "reward" are implicit, non-legible, unstructured, informal. It's about how the commun

You can't exactly decide what to reward. A lot of the important aspects of "reward" are implicit, non-legible, unstructured, informal. It's about how the community perceives you. About how many people walk up to you to chat with you at a conference. About how many requests you get to visit another lab. It's about intangible "status" and respect within the community. And it so happens that this informal respect is mostly given to people who make exciting discoveries. You can't top-down force the criteria for this. Data collection doesn't make you an exciting personality to invite to give a keynote speech or something. And those opportunities lead to more visibility, more collaborations, more papers, more grants, more recognition, more fame etc.