Been thinking more about open table ACKS lately. Stumbled on this old post and read through some of the session reports from the Shieldlands campaign, that best of open-table ACKS games.
I think the session needs to be a relevant unit of time.
Markets refresh per session (roughly).
Spells refresh per session (roughly).
Domains provide passive income per session and no XP (and may be a source of trouble; clocks can run in them, and they're part of the world in the structure of "session begins, players attempt to do a thing, pizza break, a thing happens in the world")
More explicit structure for sessions - Market Phase, Planning Phase, Adventure!, Treasure Phase
Going to need to scale things down a little from ACKS' defaults. I'm OK with a 9th level fighter having 30 mercenaries and a run-down fort - this is exactly what you'd expect from the Men, Brigands monster entry. So if you're 9th level and you have a wee baby domain that nets you 250gp/session and a handful of mercs, that's OK by me.
I've still been kicking around how to simplify mercenary bookkeeping. I think I may have hit a potential solution.
First, hire them as units beneath a sergeant, like I've been saying for a while.
Second, pay them up front.
ACKS has this concept of "the magic ratio", which comes up repeatedly in its economics and is tied to the rate of return on capital in fantasy Rome (bear with me). Most investments are expected to pay for themselves in about 30 months, and characters in the world are assumed to have about 30 times their monthly income in assets.
So if you want to hire a mercenary and not have to deal with tracking his monthly expenses, pay him 30 months of wages up front. Better, do it through a sergeant to handle a whole squad and if any one guy dies, the sergeant will take the pay pre-allocated to that guy and use it to pay some other poor bastard that he dragoons in the next market you pass through.
So if a light infantryman costs 6gp/mo in wages, then you can hire a squad of 6 of them for the foreseeable future for around 1100gp. And then mercenaries become sort of like a magic item - they can go on your character sheet instead of in a spreadsheet, because you don't need to pay them monthly. "The Yellow Saddles, Light Cavalry, 5/6 men". Easy. And they replenish between sessions subject to market class (possibly derandomized, because I don't want to have to roll for merc availability all the time).
And if you want them to replenish faster, we could base something on the commissioning items rules to let you raise the effective market class, and then either give fighters free virtual gold to spend on this every session (or let them write recruiting costs off on their living expenses), or give them a venturer-like ability to replenish mercenary units as if the market were one size larger. This also provides an incentive to visit markets during adventures, to heal your mercenary units.
Now, I could see some sticker shock at 1100gp for six measly light infantrymen. It's 2200 for heavy infantry, 2700 for bowmen, 5400 for light cavalry, and 10800 for heavy cavalry. Those are not small chunks of change, and my former players would probably want to be banking that money towards building a fortress. But if you change how domain acquisition works, from "build a fortress" to "domains are immutable, go capture one", then these changes complement each other - the money you aren't spending on castles can be spent on mercs, and the mercs are how you "buy" a castle. And honestly, with domains out of the spending picture - would you rather have a +1 spear, or six knights who respawn between sessions?
Ooh, and this works out beautifully - a 9th level fighter has 250,000 XP. He got about 200,000 of that from GP. 200,000 GP in permanent mercenaries is about 180 knights, which would be big for a bandit encampment but is in the realm of the reasonable (he probably spent some money on other things too).
So I don't think this is a totally crazy plan.