I don't think you appreciate how fortunate you have been.
I recently served jury duty on a civil case where the heart of it was about determining medical costs, and listened to people whose expertise was to determine what was reasonable and customary pricing for a given service tell me price ranges that spanned orders of magnitude sized differences in price. They needed access to multiple private databases (whose costs were prohibitive for a single use case) to determine these prices... but even with those databases they had to temper what they found with their own expertise as the data could be incorrect or incorrectly encoded. They also explained that prior to providing a service, you might anticipate only one item to be charged, the end result might actually be an order of magnitude more items charged, some of which were literally additional instances of the anticipated services.
A few years ago, when it was rolled out, I was trying to help my mother-in-law shop around for Medicare Part D plans. This should have been the easiest thing to price out: I had a list of prescription items, I knew exactly what doses and medicines she needed, and how often. I got a list of eligible providers, and proceeded to determine their their pricing and coverage rules. I figured it should be as simple as building a spreadsheet and doing basic linear optimization. First, it turned out to be very hard to get pricing, particularly in an easy to digest fashion. Beyond that, I was told the pricing could be missing some entries, could be incorrect, or could be out of date and/or could change in the near future. I also found that coverage policy invariably had subjective elements that made outcomes nondeterministic. Nobody was willing to stand behind a fixed price.
Most of the time it is a #@$@#%ing crap shoot.