One thing I've tried is capturing the min voltage during a start, with a DMM that has a setting to capture min voltage. Below is from an 18 month old Deka X2 Group 35 battery, living in So Cal.
Charged up battery, let sit overnight, then put my Agilent DMM onto it and captured the min voltage at the terminals when starting.
Ambient temp: 60F (overnight: 55F).
12.673V before test (driver door open); 10.581 min voltage captured.
Documenting a good test, then, over time, what it'll look like when questionable.
This type of test isn't one a parts counter person can do, but is a nice data point to have on a personally-owned car. To me, when the starter motor was stationary, drawing the most amps, the min voltage still was decent. A "bad battery" to me is usually one insufficient to start the car.
As far as impedance, I've seen 4.666milliohms on a new battery, 5.77milliohms on an 70% life left one.
There was one bad battery I had that measured:
11.63V, 77.35milliohms, 70°F, 1 hour after p.m. drive home.
Yet, it would quickly taper off when put on a charger.
Then it measured:
12.78V, 10.45milliohms, 269.1 measured CCA, < 40% capacity left.
I think the higher the impedance the lower the starting current (least it works that way in supercapacitors), so you want a low number. But, as I said in another thread, I don't have
much confidence in these things being able to call a battery good or bad.
Decades of experience, and a few dead chickens, are helpful.
I think if I had a submarine full of batteries I'd be doing monthly tests and graphing, looking for trends.