Daily versus lifetime budgets

A daily budget commits to a rate and leaves the total open. You keep the pacing decision — how much per day is yours — and you hand over nothing. The total spent depends on how long the campaign runs, which is a thing you control by other means.

A lifetime budget commits to a total and hands the pacing over. You state what the flight may cost and the platform decides how to distribute it across the days, typically weighting towards when it thinks the money works hardest. The rate on any given day is not yours.

That is the whole difference, and it explains everything else. Daily is a rate with an open total; lifetime is a total with a delegated rate.

Daily fails at the boundaries of a window. A campaign that must spend a fixed amount by a date has to be managed to that date by hand, and the failure mode is the month-end scramble: three weeks of steady delivery and a final week of raises that arrive too late to be spent well. That is a scheduling problem masquerading as a bidding one, and it has its own guide in Pace a budget without a month-end scramble.

Lifetime fails when the flight is wrong. The total is spread over the dates you gave, so the dates are load-bearing in a way people forget. Shorten a flight without touching the total and the daily rate rises; lengthen it and delivery thins. The number you typed has not changed and the campaign behaves completely differently.

Lifetime also fails when you want to intervene. Changing a lifetime total mid-flight changes the remaining pacing in a way that is hard to reason about, because what is left is the total minus what has gone, redistributed over what remains. Daily budgets are easier to steer for exactly this reason.

This one has consequences in the product, so it is worth stating precisely.

A daily fifty and a lifetime fifty are different money. Comparing them is a category error, not an arithmetic one, and nothing here pretends otherwise.

So where a budget change also changes the mode, the direction of the move cannot be read. The evaluator that decides whether a rule you wrote can clear a change looks for the entity's previous value and compares like with like; a changed mode means there is nothing comparable, and "cannot tell" is a refusal rather than a pass at a gate that spends money.

The practical upshot: a rule that clears budget decreases will never clear a mode change, and a mode change always waits for a person. The same applies where there is no previous value at all, or where the payload moves a field the account's mirror never held.

Both are checked against the same limits, which are stated in daily terms, so a lifetime budget is checked as its daily equivalent over the flight it states.

A lifetime amount with no readable flight is checked as-is. That is the tight reading and it is deliberate: a large total with no dates looks exactly like a large daily figure, and assuming it will be spread would be assuming the thing that has not been said.

The limits themselves. A draft may not set a daily budget above the account's own ceiling, derived from what that account has actually done — three times the highest single-campaign daily spend in its last twenty-eight days, floored at a default stated in the account's own currency. A budget may not move more than a hundred per cent of its current value in one change. An entity with no budget yet is held to a first-budget cap instead. And applied increases on one connection are summed over a rolling twenty-four hours against the account's ceiling.

None of it converts. Every comparison happens in the account's own currency, caps are stated in it, and no rate is applied anywhere on the writing side.

Stated once: your connections read your accounts, a change is written as a draft on the record with the state it would change from and the checks it passed, applying it is gated on approval — by you, or by a rule you wrote — and no platform write reaches a customer account until that platform has granted write access. Applying a cleared change to a platform is in private beta.

Daily, always, and the reason is a good argument for how to think about the two.

A plan item carries a whole-flight total — the plan is a statement about a flight, and the total at the top of the page has to be a sum of comparable things. Materialising the plan derives the daily figure by dividing that total by the inclusive length of the flight, and the flight days become instants in the ad account's own timezone with the stop exclusive.

So the division happens once, in code, rather than in somebody's head twice. It also means the plan's arithmetic is checkable: a twenty-day line carrying the same total as a twenty-eight-day line produces a visibly higher daily rate, and that is the number that meets the ceiling.

If you want the derived figure before drafting anything, the division is the whole calculation and the CPM calculator is the next one along if the question is what the rate buys in reach.

Use lifetime when the commitment is genuinely a total and the dates are genuinely fixed — a flight, a sponsorship, a fixed-budget test. Use daily when the commitment is a rate, when you expect to steer, or when the end date is a guess. When in doubt use daily, because it is the one you can change without having to reason about redistribution.

The tradeoffs of campaign budget optimisation is the next question, which is where the budget sits rather than what kind it is.