how would i actually measure my own slippage?

read the beginner guide on slippage this morning, the one where they re-price the same set of trades under a bunch of different fill assumptions. i liked it because it didn't just define the term and walk away, it made the point that the size is what matters and the size is the part nobody checks. and the gap between the friendliest and harshest assumption was a lot wider than i'd have guessed going in. so here's my actual question, and i'm asking honestly, not rhetorically: how would i test this on my own fills? what i think i can do is export my order history, and for each fill compare the price i got against... something. that something is where i get stuck. against the last trade price when the signal fired? i don't have that stored. against the mid at the time of the order? i don't have that either unless i start recording it going forward. i can compare fill price to the close of that bar but that feels like it measures a different thing entirely, it'd be mixing in whatever the price did after i decided rather than the cost of me deciding. the guide splits it into crossing the spread, walking the book, and latency, and says the first isn't strictly slippage since it's quotable up front. fine, but from where i sit as a retail account all three arrive as one number on my confirmation. is there any practical way to separate them after the fact, or do i just accept the blended figure and check whether my strategy survives it? second thing i'm unsure about. if i only ever trade a handful of large liquid names, is walking the book basically noise for my size, or is that the kind of assumption that quietly kills a backtest? i'd rather test it than assume it, i just don't know what the test looks like. not asking anyone to build it for me. mostly asking whether the sensible move is to start logging mid price at signal time from now on and revisit this in a few months with real data, instead of trying to reconstruct history i never saved.