I think part of this is if you've ever been to a foreign country, and you perform a mental calculation to make sure you aren't being ripped off between your currency and the one you now need to use. After a day or two, you barely think twice about it (if they use a "metric" form of currency, that is; I can only imagine the hijinks of visiting the UK before their currency changes for 1970).
It is actually not that difficult. 4 farthings to a penny, 12 pennies to a shilling, 20 shillings to the pound. Things ceased to be priced in farthings several decades before decimalization, so the system had been simplified even further. In practice only fairly major purchases were priced out in full in pounds, shillings, and pence. Most other things, like groceries and paperback books, were priced just in shillings and pence (often separated with a slash and a hyphen for whole shilling amounts: e.g., 5/-). The pound was not really pocket money until after inflation in the 1960's and 1970's. To give an idea of how valuable the pound was, in the 1930's £500 was a typical price for a house. I also don't think £1 was the lowest denomination of paper currency until relatively late (I believe five-shilling and ten-shilling banknotes were issued at various points in the pre-decimalization period).
The real difficulty came immediately after decimalization, when pre-decimal currency was still in circulation and people had to memorize "old money" and "new money" conversion factors until it was withdrawn from circulation (I think it has since been demonetized). As an example, two old sixpence coins would have been worth five "new pence" (two sixpence = 1 shilling = 1/20 of £1 = 5p). Occasionally, in works of popular history, you will see people do these conversions in order to make pre-decimal money amounts accessible to readers who were not born in 1970. I did not do this in my own graduate thesis, and instead quoted amounts in £
s.
d., Reichsmarks, and pre-World War II Italian lire as they were given in the original sources, but whenever I needed to give an impression of how large an amount was compared to other prospective uses of the money, I mentioned things like the cost of milk for schoolchildren, the cost of a "corporation" house, etc.
Even if we did a massive conversion, I don't think there would be a massive improvement in efficiency with business, government, and industry just because we suddently went metric; if anything, we'd be collectively flummoxed for a year or two. There would be so many holdouts and exceptions granted (sports, traditions, non-profits, as to make things even more annoying).
I tend to agree that we have already gone to metric in nearly all of those areas where switching costs would be liquidated by operating efficiencies over time. Distance indications on highway signs are not one of them, though a case could be made for height and regulated mass indications on signs. (In Britain, the regulatory basis for lorry weights has been "maximum authorised mass," expressed in metric units, since 1981, and dual-units signing for bridge clearances has been available since the late 1980's. In both cases these changes are designed to accommodate international freight movements. It has since been suggested that dual-unit clearance signing should be made mandatory, or alternatively that clearances should be signed
only in metric, since even a mild bridge strike can result in derailment.)
In the US metric units on signs are not even on the table. The
real battle, which has been raging since the mid-1990's and has now been settled decisively in favor of English units, is metrication of highway design and construction. Many highway engineers would far prefer to work in metric because units for things like stress and strain are dimensionally correct, unlike the English-unit equivalents. However, the contracting community has resisted metrication because it is costly to change out standardized components like sewer pipe and day laborers have little vernacular knowledge of metric.