Looking only at the first image, it made me comment the following: “the units of measures used in the formulas must agree, which doesn’t seem to be the case in the formula you considered. One cannot mix like this currency with the number of something.”

Otherwise, Data Downtime should be renamed to Cost of Data Downtime or something similar.

Shouldn’t the Data Downtime also consider the cost of not having the data, which can be important in decision making and can be based on SLAs, including penalties. Probably the SLAs/OLAs help to differentiate between the number of users affected by this.

Do you include the effort for testing in “resolve” component?

Conversely, data downtime could be left out or should apply only for certain products. Does it make sense to track all this information for a product that cost maybe 100 EUR?

Frankly, the more complex the metrics the less usable they are. The question is how much such a metric reflects and quantifies the reality, respectively what value does it brings? You lost me at “computing the return against a random or average decision”.

Isn’t more valuable to provide the components as metrics?

I appreciate your effort in trying to clarify all this, though if you want to sell an idea, ideally it should be done in fewer words.