Yes, and before you read another comparison, because every comparison you are about to read needs a figure only you can produce. An ownership comparison has three lines - what you pay to buy the thing, what you pay to keep it supplied, and how fast you get through the supply - and the first two are published while the third is not, anywhere, for your usage. That third line is missing because it is not a property of the product. It is a property of you: how often you get into the car, how many times you press before setting off, whether the windows are down, whether the car stands in the sun, whether somebody else drives it at the weekend. A manufacturer cannot know that and neither can I. What I can do is hand you the instrument, and the instrument is a marker pen. Step one. Mark the level tonight. Not tomorrow, tonight, while the bottle is in your hand and the thought is live. Draw a fine permanent line on the bottle itself at the top of the liquid. Put it on the glass or the plastic rather than on the label, because labels slide and peel. Step two. Write the date beside the mark. This is the step readers skip and it is the step that makes the whole thing work. A mark on its own says the level has fallen, which you knew. A mark with a date says it fell that far in that many days, and that is a rate - the thing you were actually looking for. Step three. Look again in a fortnight, and mark and date it again. A fortnight crosses a working week and a weekend, which you drive differently, and is short enough that you will not forget. Step four, and it only applies if you are measuring a refill rather than a bottle. A refill is opaque or sealed in a way that makes a level mark impossible, so the reading changes shape: note the date you opened it, and note the setting it runs on. Both halves, always. A refill that empties while running at the top setting and the same refill emptying at the lowest are two different measurements wearing the same label, and a note that records only the dates has thrown away the variable that matters most. Change the setting mid-way and you have started a new reading, so date that too. And run one count alongside the marks, because the marks alone cannot tell you why. For seven days, tally your sprays per journey. Tally, do not estimate, with a card in the door pocket and a stroke every time you press - estimated counts here are wrong in a predictable direction, because the second and third press of a single journey are not separately remembered. The marks tell you how fast the bottle empties. The count tells you how much of that emptying is your hand. Then read the two together, which is where most of the value is. A level that moved a long way with a modest count means you are a heavy, regular user of a consumable, and the response is to buy the same thing more cheaply rather than to change systems. A level that barely moved, on a bottle you were about to replace, means you were never running out and the purchases were abandonments rather than depletions. And a count that climbs through the week, three presses on Friday where there was one on Monday, is the most interesting reading of the three, because it is you topping up something you have stopped registering. What the protocol will not give you, plainly. Not a figure you can compare with a stranger, because no published average exists. Not a month at which hardware becomes correct, because the other half of that sum needs the device's own consumption and that is unpublished too. And not the decision itself. It gives you one figure that is actually yours in a category where every other figure on offer belongs to somebody's marketing department, and in my experience that one changes more minds than all the arithmetic written around it.