it shouldn't be there. Unfortunately, it rarely happens that you have a clear cut case if a feature is used or not. In most cases you see some traction, but not enough to be certain that the feature is valued by the users. One thing you can do in such situations is to kill the feature and wait for the response of the users. If no one complains, you can be pretty sure that users didn't appreciate the feature. But if they do complain, the user feedback will induce a fresh wave of enthusiasm about the feature within the product team and push the feature forward. Additionally, you can ask users advocating for the return of the feature, to participate in its development thus giving additional boost to product development.
Of course, such approach should be seen as a last resort and only used if you just cannot define the right usage metrics. If you keep on suspending and reinstalling features, even early adopters of your product will just abandon it altogether. But you also shouldn't forget that the primary goal of a start-up is not to please each and every customer, but discover a product that customers would love. Therefore, killing a feature should be one of the more important tools of a product developer.