Default to playing out weeks that still pay you. Once the milestone target is met, keep building and triggering while your engine or a useful placement produces real value. End early when the remaining weeks cannot give you anything you need and the early payout funds a concrete next step — a specific shop purchase, Council offer or cleanup. Unused weeks are not banked as spare time, so the choice is “more triggers now” against “resources sooner”.
Confirm the milestone is actually met
The milestone panel shows the score target and the weeks left to reach it. Only compare options after the target reads as complete; an engine that is one trigger short of the target is a different problem from an engine choosing what to do with spare weeks.
Build, Skip and End Early are three different moves
The week track at the top of the screen is the resource you are spending. Building keeps the run’s normal rhythm and lets a working engine keep firing. Skip, once it is open to you, passes a choice without ending the milestone — useful when the offered picks are bad but your engine still has triggers coming. End Early finishes the milestone now and takes whatever the early exit gives you. Treating a bad draft as automatic grounds for End Early skips the middle option: players in the milestone discussion point out that a Skip can still preserve valuable trigger weeks.
The default: play out weeks that still pay
Several players in the milestone thread land on the same editorial rule: buildings and triggers are usually worth more than the early Gold, because the engine you already placed keeps compounding while weeks remain. Continue when any of these is true — then reconsider.
Decision fork · conditions, not a payoff formula
Concrete “keep playing” signs: your scorer still triggers every cycle; you are one placement away from completing a chain like the ones in the herb engine or forest build; a cooldown ticks next week; or you still hold consumables that only make sense before the milestone rolls over.
When ending early is the better call
- The payout has a named job: players describe ending early to afford a specific Council offer, shop pack or tool such as a hammer or dismiss — not “Gold feels nice”, but a purchase already on the screen.
- Cleanup before the next stage: removing a hazard or dead tile is sometimes worth more than one more scoring cycle.
- Saving a big turn for later: one player lines up a high-value play for the next milestone’s opening instead of spending it on a target already met.
- The engine is done: picks are bad, nothing in range improves, and every extra week is genuinely dead time.
Two cautions keep this honest. First, players report exceptions — a specific heirloom pairing was cited as flipping the usual math, so check your own heirlooms before applying any rule. Second, nobody in the thread has a verified Gold-per-week or XP formula: this is a judgement call with conditions, not arithmetic. If missing cards are the real reason your weeks feel dead, read why a card never shows up before blaming the milestone, and see the guild pair plans for what a working engine looks like.
The decision cases above come from the Ending Milestones early discussion (Sep 21–Oct 1, 2026, several authors) and emgeal’s Sep 6 interface guide, read Oct 5, 2026. They are attributed player experiences and our editorial synthesis — not an official optimal rule or a current-client formula.