While I agree with your criticism it boils down to the fact that measuring the impact of lobbying is difficult. I haven't seen convincing evidence either way as to whether the ROI is high. I suspect it is, at least for sophisticated firms, but is capacity constrained: You can't spend money if there isn't an issue to lobby and any given issue is subject to diminishing returns.

That would be consistent with high-ROI but limited revenue in lobbying.