So in this scenario, say you don't know what the end user is planning to do with the collection of reports and that the user themselves might not have a clear idea of ROI until the report is in their hands and they are able to extract some value from it. Simply having more of it does not necessarily guarantee value but rather what they do afterwards that determine the value of the reports and the quality of the reports (which they choose the data source).

Does this make sense? Basically the issue I'm raising is, what if you can't predict a positive ROI from the reports generated by the user of your software until their intentions with the report is known and after they take action?

...but someone will argue the act of generating the reports automatically and the volume of reports generated IS the value. This argument also has merit because if they were to hire someone to do this, it would cost them X amount of money and using software saves them X amount of dollar and time. The quality of the data source they chose to generate the reports is outside of the software's control.

Maybe I'm just overthinking this...but it does seem like metered billing makes sense because in order to generate the same number of reports, it would cost time and money. If somebody determines that they can do it better and faster and cheaper but still wants to use the software and not pay for even 1/10 of the cost of hiring someone to do it, then maybe they are a niche to not pay attention to. Maybe there's no money in that niche, and it should be ignored altogether. Perhaps even targeting this niche that is resistant to paying for even 1/10 of the cost of generating reports without the software should not play a factor in pricing.