Google Ads marks a campaign Limited by budget. Its Target CPA is $100, while its mature actual cost per conversion is $65. Before changing the target, compare both with the business’s acquisition allowance under the same conversion definition. A gap deserves investigation; it does not identify a profitable replacement target.
Google’s documentation explains the bidding change. It cannot predict this campaign’s result or prove that a target adjustment will increase paid orders or profit. Target CPA is a desired average cost per conversion, not a ceiling on every order. Google’s Target CPA guide makes that distinction explicit.
Confirm that this change applies
Google says the rollout completed on August 27, 2026. It affects qualifying budget-limited campaigns using Target CPA, Target ROAS, or Target CPC for Demand Gen. The named scope includes Search, Shopping, Performance Max, Demand Gen and Travel; management surfaces include Google Ads and Search Ads 360, plus Demand Gen in Display & Video 360. Google does not automatically change targets or daily budgets. See its change notice for the scope.
Check the budget type and the real owner of the setting. Google’s FAQ excludes campaign total budgets from this behavior change and explains how portfolio strategies and shared budgets are handled. An individual campaign view may not contain the control your team needs to review. If the campaign is outside the documented scope, make no change on the strength of this announcement alone.
Put the target, the result and the allowance side by side
Each number has a different job. The target tells the bidding system what average result to pursue. Actual CPA describes what happened under the selected conversion definition. The commercial allowance describes what the business can afford to pay for an order or customer.
For a historical comparison, use the average target for that period, not just the setting visible today. Target changes and ad-group adjustments can make those numbers differ. The bid strategy report helps compare the target with actual performance and assess conversion delay.
Before translating CPA into order economics, check what counts as a conversion. Purchases, leads and other actions cannot share one order-cost assumption. Nor can you divide a campaign’s spend by every order in Shopify and call the result that campaign’s acquisition cost. Declare a defensible order or customer scope, and keep any attribution uncertainty visible.
Build the allowance from contribution before advertising: sales after the relevant discounts, returns and variable costs. Set aside the amount needed for profit and other business costs. Where repeat purchases are part of the allowance, state the payback period and evidence for them. Do not quietly turn an uncertain future value into cash available today.
A fixed budget can still buy a different result
Consider a deliberately simplified example. Assume one measured conversion corresponds to one comparable paid order, with the same period and currency. The figures are illustrative, not a forecast.
At the same $6,500 media spend | Arithmetic |
|---|---|
100 conversions | $65 average CPA |
65 conversions | $100 average CPA |
$80 contribution before ads per order, with a $10 reserve | $70 acquisition allowance per order |
Under those assumptions, 100 orders contribute $8,000 before media. Subtracting $6,500 leaves $1,500 before the reserve. At 65 orders, $5,200 of contribution leaves a $1,300 shortfall after media, before the reserve.
This calculation does not say Google will produce either outcome. It shows why keeping spend fixed does not make a $100 target commercially harmless. It also shows why a $65 result alone cannot select the next target: the allowance, reliability of the result and possible loss of delivery all matter.
If the conversion is a lead, the order mix changes, or returns erase the assumed contribution, the example no longer fits. Replace its assumptions before using its arithmetic. Neither scenario measures incremental orders caused by advertising.
Complete the target-to-margin card
Use one record for the campaign or shared control being reviewed. Keep the inputs beside the decision so another operator can understand it later.
Record | What to write down |
|---|---|
Exposure and owner | Campaign type, strategy, budget type, Limited by budget status, and who owns any shared control |
Data and dates | Conversion actions, value definition, currency, date basis and a period mature enough for the decision |
Target and actual | The period-average target and actual CPA or ROAS on that same definition |
Commercial allowance | Order or customer scope, contribution before media, reserve and acceptable acquisition cost or return |
Choice | Keep the target, align it deliberately, investigate first, or make an explicit strategy change |
Next check | Budget range, other changes, conversion maturity, check-in date and the result that would make you stop or reconsider |
Keep the target when it still represents the average outcome the business wants and the measurement is usable. Recent performance better than the target is not a guaranteed baseline.
Align the target deliberately when the saved setting permits economics outside the business’s allowance. Use mature evidence to choose the number, and accept that a tighter target can reduce spend or volume. A suggested target in the account is an input to that decision, not proof of profitability.
Investigate first when a tracking fault, changed goal, unresolved returns or a short reporting window makes the comparison unreliable. Low conversion volume is a reason to widen the uncertainty, not to pretend the latest CPA is a stable estimate.
Change strategy deliberately when the actual objective is to maximize volume or value within the budget without the current efficiency target. Removing the target changes the instruction. Record that tradeoff and its stop rule instead of using it simply to clear a warning.
Do not apply an order-cost rule to every target
For Target ROAS, inspect the value sent to Google. Gross attributed revenue is not contribution margin; changes in discounts, product mix and returns can alter the commercial meaning of the same reported return.
For Target CPC, the immediate objective is clicks. Keep click cost as a delivery diagnostic and examine paid-order economics separately. A $1 click target cannot be compared directly with a $70 acquisition allowance.
Consent choices, duplicate events or a changed purchase definition can also move reported conversions without creating or losing orders. Resolve those measurement questions before using the apparent change to justify a new bidding target.
Change one decision, then read a mature result
Save the prior target and the completed card. Keep the budget range and other controllable inputs stable enough to interpret the change, and record changes in creative, offer, stock, prices or conversion goals. A fixed daily budget setting alone does not establish comparable realized spend.
Google’s FAQ recommends allowing one to two conversion cycles when evaluating the change. Use the account’s actual delay; an immature result can make CPA look worse before conversions arrive. A planned wait does not override a broken tracking path or an agreed commercial stop condition.
If performance moves, keep competing explanations in view: demand or auction conditions, delayed or redefined conversions, and changed customer or product economics. An uncontrolled before-and-after comparison cannot tell you how much of the movement the bidding change caused.
Leave the review with a decision another person can act on: the target to keep or change, the commercial allowance behind it, the next mature check and the result that would reverse the choice. If you cannot complete those fields, the next action is to resolve the missing input.
