> ## Content Index
> Fetch the complete content index at: https://cascader.io/blog/llms.txt
> Use this file to discover other available public pages before exploring further.

# Before the sale, what should change in Google Ads?
- URL: https://cascader.io/blog/before-the-sale-google-ads-plan/
- Published: 2026-10-06T12:00:00.000Z
- Updated: 2026-10-07T18:54:59.000Z
- Description: Three worked promotion plans separate conversion-rate changes, order economics and funding—and show what to restore when the sale ends.
- Author: Rick Hecker
- Tags: #Import 2026-10-05 03:19

[Cascader](https://cascader.io/?ref=cascader.io)

“We’re doing 20% off this weekend. Can you get Google Ads ready?”

You can. First, the business needs to supply rather more than the percentage on the sale banner. More people shopping, more shoppers buying, and less money left from each order are three different changes. They can happen together. They can also pull your account settings in opposite directions.

The useful preparation is a sale plan: what you expect to change, why you believe it, what an order can afford to pay for advertising, how much cash and stock you can commit, and exactly how the account comes out of the promotion. The following three worked plans show why that conversation should happen before anyone reaches for an adjustment.

## Start with the order, then the traffic

Ask the business to price a representative sale order after the discount. Subtract the costs that follow that order: goods, fulfilment, payment fees, shipping subsidy and a realistic allowance for returns. The remainder is contribution before advertising. Decide how much of it must remain after advertising to help cover overhead and profit.

**Allowable acquisition cost = contribution before ads − contribution the business needs to retain.**

That is a commercial allowance, not a promise that bidding will achieve it. Use a realistic basket and product mix; a campaign selling both clearance stock and full-price bundles cannot be governed safely by the margin on its nicest item. If the business deliberately accepts less contribution to clear stock, write down that choice and its limit. If it wants to justify a loss with repeat purchases, ask for the repeat-purchase evidence and the cash to finance the wait.

Next, separate the traffic forecast from the conversion-rate forecast. A previous sale producing 50% more orders tells you little by itself: it might have bought 50% more clicks at the same conversion rate. Compare clicks and purchases in the affected campaigns, using conversion data that has had time to arrive. Check whether the offer, audience, devices, products and checkout were comparable. An email blast that changes the audience mix is a reason to question the comparison.

Finish the brief with stock available to paid traffic after other channels, fulfilment capacity, approved spend across the affected campaigns, and the offer’s start and end with a named time zone. Include who can stop it. “We’ll keep an eye on it” is often an excellent way to discover that everybody thought somebody else was keeping an eye on it.

## Plan one: the sale converts better, but each order can pay less

Imagine Ember, a cookware shop running one established Search campaign with revenue-based target ROAS bidding. Its three-day sale runs from Friday 00:00 until Monday 00:00 in the account’s time zone. Sale prices stop at checkout at that boundary; there is no coupon that stays valid afterwards. These businesses and figures are illustrative.

The business expects its usual $100 basket to sell for $80\. Costs remain $60 per order. Finance wants at least $10 left after advertising.

| Per order                             | Ordinary price | Sale price |
| ------------------------------------- | -------------- | ---------- |
| Revenue                               | $100           | $80        |
| Costs before advertising              | $60            | $60        |
| Contribution before advertising       | $40            | $20        |
| Contribution to retain                | $10            | $10        |
| Allowable acquisition cost            | $30            | $10        |
| Revenue ROAS needed at that allowance | 333%           | 800%       |

The 20% discount has cut the advertising allowance by two-thirds. This is where the cheerful instruction to “lower ROAS for the sale” meets the person who pays the invoices.

Ember’s ordinary target is 400%. At $100 revenue per order, achieving that return means spending $25 to acquire an order, leaving $15 contribution. At an $80 sale price, the same 400% return permits $20 acquisition cost—every dollar of the contribution before ads. Revenue tracking can report the lower price correctly and still leave bidding with the wrong commercial goal.

For this sale, Ember chooses an 800% revenue ROAS target: $80 ÷ $10 × 100\. That is the minimum return consistent with its allowance in this simplified basket. A business needing a buffer would choose a stricter goal or fund less exposure. Target ROAS optimizes against the conversion values supplied to Google Ads; it aims for an average return, so setting 800% does not guarantee every order, or the event overall, meets it. [Google’s target ROAS documentation](https://support.google.com/google-ads/answer/6268637?hl=en&ref=cascader.io) explains that distinction.

If your account sends contribution rather than revenue, use that value in the ratio. Ember’s $20 contribution divided by $10 advertising allowance would be 200%, not 800%. Those two targets describe the same economics on different measurement bases. Confirm what the account actually sends; [conversion values can represent revenue or profit](https://support.google.com/google-ads/answer/13064207?hl=en&ref=cascader.io). A sale weekend is a poor occasion to discover that the column called “value” means something different to finance.

Ember also has evidence for a conversion-rate lift. Two comparable promotions, after conversions matured, moved purchase conversion rate from about 2% to 3% in this campaign. The coming offer and product mix are similar. The working expectation is therefore a **50% relative increase**: 3% ÷ 2% − 1\. It is a one-percentage-point increase, but 50% is the number for the adjustment.

Ember schedules a conversion-rate seasonality adjustment of +50% for the affected campaign during those three days. This informs bidding about the expected likelihood of a purchase. It does not set a 50% bid increase, grant extra budget or make an $80 order worth $100.

The two bidding changes are coherent: shoppers are expected to buy more readily, while the business can afford less advertising per purchase. For intuition, allowable acquisition cost × conversion rate gives a simple affordable cost-per-click estimate. Ember moves from $30 × 2% = $0.60 to $10 × 3% = $0.30\. This is planning arithmetic, not an auction bid instruction. Even with the better conversion rate, its economics support a lower click cost. The operator must accept that the sale may buy less traffic than the banner’s owner hoped.

Ember keeps its $300 average daily budget. At an illustrative $0.30 CPC, 3,000 clicks over three days would produce 90 orders at 3% conversion rate: $7,200 revenue, $900 spend and $900 contribution after ads. That is a forecast to test, not an entitlement to those clicks. There are 180 orders’ worth of stock allocated to paid search after other channels. Finance approves funding for possible spend above the $900 forecast, including up to $600 on a day under the usual charging limit for most campaigns: twice Ember’s unchanged $300 average daily budget. Finance also checks that the event fits the month’s cash allocation.

The operator changes the target when the sale economics begin, verifies the scheduled conversion-rate adjustment, and leaves the budget alone. During the event, the owner checks spend and remaining paid-search stock, tests checkout, and examines sufficiently mature orders against the $10 allowance. If the offer breaks or the stock allocation is exhausted, the affected advertising stops. If mature acquisition cost makes the retained contribution unacceptable, Ember reduces exposure or ends the paid promotion rather than lowering the return requirement to make the campaign spend.

At Monday 00:00, the conversion-rate adjustment ends automatically. The operator verifies that checkout prices are back to normal and restores the 400% target. The $300 budget never changed. Sale-period clicks still need time to finish converting before the event is judged.

![A continuous band beside Ember’s dated changes shows the $300 average daily budget unchanged throughout. Friday through Sunday: a +50% conversion-rate signal and 800% revenue ROAS target. Monday: the signal expires automatically; after normal prices are verified, the target returns to 400%. The results review follows conversion lag.](https://cdn.mymidnight.blog/ced556cd9f9c0c8315cfbe0744a3baf0/2026/10/cascader-20261004-idea-002-sale-clocks-v02-1.svg)

The operational ending and the analytical ending are separate. Restore settings when the conditions for new traffic change. Review the sale’s results when the sale’s conversions have arrived.

## The two seasonal adjustments do different jobs

One adjustment tells bidding about an expected change in purchase probability. The other schedules extra funding. Neither decides what the business can afford to pay for an order.

Google’s **conversion-rate seasonality adjustment** supplies an expected major change in conversion rate. Smart Bidding already handles ordinary seasonality; this tool is for exceptional short events, ideally one to seven days. Google warns that periods beyond 14 days may work less well.

The conversion-rate adjustment supports Search, Shopping and Display with target CPA or target ROAS. It also supports Performance Max and App campaigns with all bidding strategies; App support is beta. Travel is excluded. The scheduled signal ends without a compensating negative adjustment. [Google’s seasonality guidance](https://support.google.com/google-ads/answer/10369906?hl=en&ref=cascader.io).

Google’s **seasonal budget adjustment** schedules an additional amount on top of the average daily budget. It supports three to 14 days on eligible Search and Shopping campaigns.

The budget adjustment excludes shared budgets, dayparting, flighted campaigns, drafts and conflicting scheduled budget adjustments. Performance Max is not listed as eligible. At the end, the budget returns to its prior value, although intervening manual changes can affect that return. [Google’s budget-adjustment guidance](https://support.google.com/google-ads/answer/12922263?hl=en&ref=cascader.io).

## Plan two: fund more demand, with no conversion-rate adjustment

Harbour, a homewares retailer, runs a five-day bonus-bundle promotion through an eligible Shopping campaign. The customer still pays $60 for the average order. The supplier pays for the bonus item and its added packing, leaving $25 contribution before ads unchanged. Finance requires $10 retained contribution, leaving a $15 acquisition allowance. The existing 400% revenue ROAS target matches that allowance: $60 ÷ $15 × 100.

Comparable promotions brought more paid-search traffic while purchase conversion rate remained near 2%. Recent comparable periods also show the campaign running out of budget while delivering acceptable mature returns. The plan anticipates 1,500 available clicks a day at an illustrative $0.20 CPC, rather than the usual 1,000\. It expects 30 orders a day rather than 20\. The evidence supports buying more traffic, not telling bidding that each click is more likely to convert.

Harbour leaves its 400% target alone and creates a five-day seasonal budget adjustment of **+$100**, taking the average daily budget from $200 to $300\. The working event forecast is $1,500 spend for 150 orders, $9,000 revenue and $2,250 contribution after ads. The supplier funding must actually be collectable; an unsigned promise is not a margin.

The business allocates 300 orders to paid traffic and approves cash headroom above the forecast, accounting for possible $600 spending days and the revised monthly limit. One owner checks spend, stock and whether the additional traffic retains the expected economics each day. If demand fails to materialize, there is no duty to spend the forecast. If the new clicks convert worse, the owner revisits their quality and the funding decision; “more demand” has not licensed worse economics.

At the scheduled end, the owner verifies that the budget returned to $200 and checks that the supplier offer, ad copy and product prices have ended together. There is no conversion-rate signal to reverse and no target change to undo. The results review includes the supplier payment and returns, not merely the Google Ads revenue column.

## Plan three: prepare the promotion and leave bidding alone

Rowan, a stationery shop, offers a free gift on Saturday and Sunday. The supplier pays for it, packing capacity is available, and the gift does not change the shop’s $30 contribution before advertising per order. The business requires $10 retained contribution. Its existing $18 target CPA remains inside the resulting $20 allowance.

There is no comparable evidence of a conversion-rate lift. Ordinary seasonal demand already fits inside its $180 average daily budget. The forecast is 500 clicks a day at $0.36 CPC and 2% purchase conversion rate: 20 orders and $360 spend across the weekend. The paid channel has 40 gifts allocated after other channels; the business has approved the possible spend above the forecast and knows who will respond if stock or cash approaches its limit.

Rowan changes the offer on the page and in the ads, tests gift eligibility and purchase tracking, and appoints the weekend owner. It leaves the target and budget unchanged. It does not enter a speculative conversion-rate uplift merely because somebody made a banner.

The owner watches for a real capacity problem or unexpected demand and can reopen the funding decision if either appears. At the end, they remove the gift promise and verify checkout. There are no temporary bidding settings to remove. Preparing the account has consisted of making the promise accurate, funding the existing exposure and arranging supervision. That is a complete plan.

## Put the start, stop and review on the same page

Before making changes, record the baseline settings and the approved temporary settings by campaign. Scope the plan to the traffic and products that actually experience the offer; a shop-wide announcement does not make every campaign’s conversion rate move by the same amount. Check strategy and adjustment eligibility in the account.

Then give the plan these four appointments:

1. **Before launch:** confirm the basket economics, evidence for each forecast, cash allocation and paid-channel stock. Test the discounted checkout and reported purchase value. Match the time zones and boundaries across ads, site and product feeds. Merchant Center’s sale-price effective date supports explicit time zones; omitting one defaults to UTC. [Sale-price date specification](https://support.google.com/merchants/answer/6324460?hl=en&ref=cascader.io).
2. **At launch:** the named owner verifies the actual customer offer and the live settings. If prices or checkout fail, postpone the paid plan. For manual target or budget changes, specify who makes them and who covers their absence. A scheduled intention is still worth checking.
3. **While live:** check spend and available stock often enough to act before the business limit, and distinguish those immediate constraints from delayed performance data. Decide in advance which conditions trigger a pause and which merely trigger investigation. A broken checkout or unavailable stock does not require waiting for mature ROAS.
4. **At the end, then after lag:** remove the customer promise and verify every scheduled restoration. Restore manual targets when the commercial conditions for new traffic return to normal. If the promotion was extended or buyers can still claim the discount, update the exit plan to match the real offer. Later, review the sale-period click cohort with its conversions sufficiently mature, plus actual contribution, returns and stock effects.

Budget arithmetic deserves its own line in that plan. **An average daily budget multiplied by the number of sale days is a pacing forecast, not a hard event cap.** For most campaigns, a day can be charged up to twice the highest average daily budget set that day; lowering it later does not erase that day’s higher limit. Budget changes also affect the remaining monthly allowance. Get approval for the exposure, monitor with headroom and leave room for reporting delay. If a strict event cap is essential, investigate a campaign total budget: eligible Search, Shopping and Performance Max campaigns can use one for three to 90 days, but an existing average-budget campaign cannot simply be switched to it. That is a separate campaign-design decision. [Google’s budget-change rules](https://support.google.com/google-ads/answer/10487143?hl=en&ref=cascader.io).

Targets are goals too. Google says a Search Smart Bidding target change begins affecting bids quickly, but achievement and assessment can need one or two conversion cycles. Avoid repeatedly changing the goal in response to incomplete results. [Target-change guidance](https://support.google.com/google-ads/answer/10433846?hl=en&ref=cascader.io). Conversion lag can make recent CPA look too high and ROAS too low; use the account’s lag evidence to choose the review date. [Conversion-lag reporting](https://support.google.com/google-ads/answer/9347141?hl=en&ref=cascader.io).

The final agreement should fit in a short message: “For these campaigns, this offer changes these economics; we expect this traffic and this purchase rate for these reasons; these are the targets and budgets; this is the funded exposure and stock allocation; this person checks it; these conditions stop it; these settings return at this boundary; these results are reviewed after this lag.”

That is something the business can approve and the operator can execute. “20% off this weekend” can stay on the banner.