12× the budget
in eight weeks.
Orders got cheaper.
BRIGHT Sport sells glow-in-the-dark sports balls — football, basketball, volleyball — so the game carries on after dark. Almost everything it sells, it sells at Christmas. Three markets carry the account: the United States, the United Kingdom, and Australia with New Zealand.
Between October and December 2024 the monthly budget went from €10.7K to €128.6K — twelve times the money in eight weeks. Orders went from 429 to 6,141. And the cost of an order went down, from €24.89 to €20.95.
That is the only thing this page is really about: the account got better as the load went on it, at the one time of year when there is no room to be wrong.
Month by month
Revenue and spend · Google Ads conversion value · EUROctober is the quiet month and was run that way — €10.7K of budget, 429 orders. Then the season opens. November takes seven times that budget, December twelve times, and the price of an order falls by about four euros along the way.
Cheaper orders on a much bigger budget is the opposite of what usually happens when you scale, and it is the whole reason this quarter is on the site. Most accounts pay for volume with a worse cost per order. This one got a better one.
How the quarter was built
Four decisions, all of them taken before OctoberA peak quarter is won before it starts. Nothing here was invented in December — by the time the season opened, the account was already running the shape it needed.
1 · A line per market
The US, the UK and Australia with New Zealand each ran their own search campaign instead of one campaign for the English-speaking world. Budget, bids and cut-offs move per market, so a slow country cannot quietly drain a fast one. Those three lines produced the three cheapest order costs in the quarter.
2 · A permanent layer and a seasonal one
The Collection search lines run all year. A separate Christmas layer in Performance Max was switched on for the peak and switched off after it. The season got its own budget and its own creative, and never had to borrow from the campaigns that have to work in January.
3 · December was a scale-up, not a cold start
The account was already spending €5K–€11K a month through late summer and early autumn. By the time budget went up twelve times, the structure and the bidding had months of data underneath them. Google does not learn quickly in December, so the learning was done in September.
4 · Search bought the customers, PMax bought the volume
Of the six lines that carried the quarter, search took about half the money and returned nearly 60% of the revenue, at €15.70 an order. Performance Max took the rest and did what it is good at: reach. Neither was asked to do the other's job.
Where the money went
Q4 2024 · the six lines that carried the quarter · 69% of all revenue| Campaign | Spend | Revenue | Per order | Status in the account |
|---|---|---|---|---|
| Search · US | €17.7K | €99.6K | €11.66 | Cheapest customers in the account. Capped by budget. |
| Search · UK | €29.0K | €80.1K | €18.24 | Biggest search line. Capped by budget. |
| Search · AUS & NZ | €17.5K | €64.1K | €17.84 | Capped by budget. |
| PMax Christmas · US | €14.4K | €63.6K | €14.69 | Seasonal push. Also capped by budget. |
| PMax Christmas · UK | €28.5K | €62.7K | €22.88 | Most expensive orders of the six. Not capped. |
| PMax feed · US | €17.5K | €51.0K | €20.91 | The catalogue line. Steady. |
Read the last two columns together. The three search lines bought 4,085 orders at an average of €15.70 and all three were running into their daily ceiling. The UK Christmas line bought its orders at €22.88 and had room to spare.
We ran this on the yardstick we now argue against
Those budget caps were not an oversight. They were set to hold a blended return number — the same number every agency reports and every founder checks. When the return dips, the cap comes down. That is first-order buying, and this account is our own example of it.
Nobody was being careless. A cap is the only sensible move if the only thing you know about a customer is what they spent in the first thirty days. The yardstick was the problem, not the buying. Ask a different question — what is this customer worth by day 90 — and the caps go on the other line.
What the return did, and what that tells you
Q4 2024 versus Q4 2023 · Google Ads conversion value| Month | Spend | Spend YoY | Revenue | Revenue YoY | Return |
|---|---|---|---|---|---|
| October 2024 | €10.7K | −43% | €24.1K | −62% | 2.25× |
| November 2024 | €77.4K | +52% | €220.0K | −5% | 2.84× |
| December 2024 | €128.6K | +124% | €365.8K | +89% | 2.85× |
| Whole quarter | €216.7K | +71% | €609.9K | +25% | 2.81× |
Across the year the return fell from about 3.8× to 2.81×, a drop of 26%, while the budget went up 71%. On a first-order reading that is a bad quarter. Inside the quarter the same account went 2.25× → 2.84× → 2.85× as the budget went up twelve times.
Both readings are true, and neither one answers the question that decides whether the quarter was worth doing: what did those 10,468 customers do next? Nothing on this page can tell you. That is not a gap in the reporting — it is what the metric is.
What we would do differently
Take the cap off the €11.66 line before the season opens, and pay for it out of the €22.88 one. That is a budget move worth roughly €40K of revenue at the rates the account was already achieving — and it needs no new campaign, no new creative and no new agency.
What we would measure instead
Cost per new customer against what a customer is worth by day 90, per market. Then the caps get set by the payback window rather than by a monthly average — and December's budget stops being an act of nerve.
What this doesn't prove
- These are Google's numbers, not the brand's. Revenue here is the conversion value Google reports for its own clicks. It is not the brand's total revenue and it is not incrementality-adjusted. Our pet-care page is measured the harder way; this one is not.
- No payback window. No cohort was verified, so we claim none. A return tells you nothing about margin, and nothing about what happened after the first order.
- BRIGHT would not qualify for ARQ today. It sells equipment people buy once and give at Christmas. We buy customers for brands whose customers come back. This proves execution under seasonal load — it proves nothing about the Payback Standard.
- It is dated. Q4 2024, closed and reviewed in January 2025. It is on the site because the pattern holds, not because it is recent.
- This is the founder's Icarus track record, not an ARQ client. A Meta agency ran alongside the whole time. We have never asked a client to drop one.