There are a lot of factors that go into the efficiency. I would like to depend on the numbers the car shows on the screen. True efficiency would be miles driven divided by kW dispensed. it would account for all the losses along the way. For an ICE , it’s easy. I fill up the tank and record the miles. Drive and fill up again, record the miles driven. I look at the pump for the number of gallons dispensed and calculate my mpg. of course it varies with every fill up, but I can get a sense over time as to how the car is performing through cumulative numbers. My mpg calculation is very close to what the car tells me. For an EV, some of the kW get lost along the way, unlike filling up a gas tank (unless you overflow and spill gas on the ground). Those losses need to be taken into account for true efficiency. After all, I am paying for all the electrons dispensed, not just for the ones entering the car and I can get an apples to apples comparison with my ICE.
Then there is the efficiency on the road to accurately account for range, which I also want to know. I need to know how many kW are in the battery when I start, how many miles I have driven and then how many kW it takes to fill up. The kW it takes to fill up has those cable and heat losses included so I won’t have a real idea as to how many kW entered the battery. The car computer should account for that and give me an accurate reading for miles driven divided by kW consumed. I can calculate if it is giving me an accurate picture. For arguments sake, assume the battery size is 100kW when full. I drive it down to 20% which means I have used 80 kW. If the car says I averaged 3 miles per kW, then I should have driven 240 miles. That should match the speedometer reading. It is THAT number that I want to be accurate on the car when on trips.