One of the big changes the 19th century brought in the wake of the French Revolution, was that we began to register everything. The concept of a census and people registers is an old one, but in modern history it became a standard operating measures of modern states. Everything is registered in modern states, when (and where) you are born, when (and where) you die, when (and who) you marry & everything in between and so much more.

So basically, you'd have an administrative footprint that allows you to compare.

Easily actually.

You just take how many men you SHOULD have, and see how many are missing.

You ALWAYS have a running count on these numbers, because that's your manpower.

Think of it like this:

If I have a squad of 12 men, and two of those men get killed, I have to report that, THAT DAY. At the end of every day, I am passing back status updates on how many men, weapons, ammo, supplies etc etc I have, and what I've spent.

If I am a commander, and I send a force of 1,000 to go land on a beach, at the end of the day, I am going to get a report back on what my status and remaining unit strength is.

If I had 300 men killed, 150 injured, and 50 missing, that all has to get counted up by their units, reported up the chain, and tracked by their units, and the units above them, all the way up to me.
Because I need to know that of 1,000 men I sent to that beach, right NOW I only have 500 combat able men left.

This is how I stay aware of questions like:

• the Task Force still functional? Do they still have enough strength to keep working?

• Do they have enough men left to defend the beach they just took?

• If not, what do I do about it? Do I replace the whole unit with another healthy one? Or do I just send some spare men to replace their losses? Where can I get those spare men from? What numbers does everyone else have? Do I just accept that we CAN'T hold that beach, because we don't have the numbers to do it?

Meanwhile, the analysts, and war planners are getting those same numbers too, to answer questions like

-Based on our losses at Tarawa, Peleliu, Guadalcanal, etc "approximately how many men does it cost to take an island?"

-Do we have enough men to take all the islands we need to? Is that plan possible? Is that plan acceptable? Even if we CAN sacrifice enough lives to take every island by force, are we WILLING to? Or do we need to look for an alternative plan? (see: Not invading the Japanese mainland by foot)

-What tactic/strategy changes resulted in more/less casualties? What's working, what's not working?

TL;DR:

WWII, WWI, and every war before it, as far back as civilization goes, armies have always kept running counts of deaths, because

Before a battle - "How many men do we have?"

After a battle - "How many men do we have left?"

are critical tactical/operational/strategic pieces of information. Every unit commander at every level would be tracking and reporting to the people above them.

*Note, the historical accuracy of OUR modern day estimates of those numbers will be affected by the quality and accuracy of the record keeping, and how long ago the records were taken.

*Second note, for a nice theatrical depiction of what we mean here, fast forward this video to 40:45