In this paper we investigate the effectiveness of imposing scrutiny to fight firms' information garbling. We study a setting in which both firms with good and bad projects are able to influence the informativeness of a public signal regarding their project types through unobservable efforts, and scrutiny will be imposed if the outcome of the project is bad while the previously signal about the project type was good.We show that a scrutiny threat is indeed effective in deterring bad firms' garbling behavior. However, in most cases it is optimal not to impose scrutiny at all, even if part of the scrutiny cost is reimbursable to investors. This is because scrutiny not only deters bad firms from garbling, but also hurts good firms by punishing them for "bad luck." A good firm may be accidentally punished if the signal is accurate about its type but the outcome unfortunately turns out to be bad. In most cases, the benefit of suppressing garbling is outweighed by the cost of hurting and discouraging good firms, and imposing scrutiny is not efficient. We also find that when a sufficiently large proportion of the scrutiny cost can be reimbursed to investors, a good firm's effort to improve information quality may even increase in the scrutiny cost.