There are two ways to get carbon dioxide into finished beer, and they are less different than they look. You can push it in from a cylinder, which is what kegging usually means, or you can give the yeast a measured amount of sugar and let it make the gas inside a sealed container: a bottle with a carbonation drop in it, or a keg primed exactly like a bottle. Both routes end at the same place, described by the same number: volumes of carbon dioxide dissolved in the beer. Everything else is a question of how long you are prepared to wait and what equipment is on the bench.
What a volume of CO2 actually is
One volume means the beer holds as much carbon dioxide, measured as a gas at 0°C and one atmosphere, as there is beer. Because the molar volume of an ideal gas at those conditions is 22.414 litres per mole and carbon dioxide weighs 44.0095 grams per mole, one volume works out at about 1.96 grams of dissolved CO2 per litre of beer. That is the unit every carbonation figure on this site is in, and it is why a target of 2.2 volumes in 23 litres is a definite weight of gas rather than a feeling about fizziness.
What carbonation drops are, and what they are not
A carbonation drop is a pressed tablet of sugar, sized so that one goes in a bottle. Their convenience is real and their weakness is that a drop is one size while bottles and target carbonations are not: the same drop in a 330 ml bottle and a 500 ml bottle gives noticeably different beers. If you are priming to a number rather than to habit, weigh loose sugar and use the drops for the last few bottles when you have run out of patience.
Priming a keg is priming a bottle, only larger
A sealed keg holds pressure exactly as a bottle does, so the same arithmetic applies: work out the gas the beer still needs, divide by what a gram of your sugar can make, and add that weight. The advantage over forcing is that it needs no cylinder and no regulator; the disadvantage is that it takes a fortnight at room temperature and cannot be adjusted afterwards. A keg you primed slightly flat stays slightly flat unless you have gas.
Forcing, and why the beer is never starting from zero
Setting a regulator and waiting is the fastest route, but the figure to set it to depends on the temperature the keg is held at, because cold beer dissolves more gas at the same pressure. The same physics decides how much carbon dioxide your beer is already carrying before you do anything: at the end of fermentation the beer is in equilibrium with the gas above it, and a beer that finished at 12°C holds about 1.13 volumes against about 0.81 at 24°C. Prime or force both to the same target and the warm one needs materially more.
Questions people ask about kegging
How much carbonation should I aim at?
British ales sit around 1.5 to 2.0 volumes, most pale ales and lagers 2.2 to 2.6, and wheat beers 3.0 or more. Above about 3 volumes you are into bottles designed for it; a standard crown-capped bottle is not.
Can I keg and prime rather than force?
Yes, and many brewers do when the cylinder is empty. Use the same weight of sugar you would have used for the equivalent volume in bottles, seal the keg, and leave it somewhere warm for two weeks before chilling.
Why do the drops recommend one per bottle regardless of size?
Because it is simple, and simplicity sells. It is also why bottles primed with drops vary: the same drop is a much heavier dose per litre in a 330 ml bottle than in a 500 ml one. Weigh sugar if the target matters to you.
Does the beer lose the carbonation it already has when I rack it?
Some, yes: every splash and transfer drives gas off. The residual figure this site uses assumes the beer went into the packaging container without being knocked about. If you racked roughly and the result is flat, that is where it went.