StoryKeeper turns a spoken interview into a hardcover memoir — a gift people buy for the parent who would never write their own book. Almost all of that demand lands in one window: Mother's Day. We scaled that peak across the US, Canada and Australia, on four channels, without letting the account burn efficiency chasing the spike.
StoryKeeper sells a gift, and gifts have seasons. The product is an AI-guided memoir: the recipient answers spoken questions on their phone, no typing, and StoryKeeper compiles the answers into a hardcover book. It's the kind of thing people buy for a milestone, and the biggest milestone by far is Mother's Day. A huge share of the year's revenue has to be captured inside a few weeks.
That created two problems at once. First, the brand sells globally — the US, Canada and Australia each behave differently, sit in different currencies, and hit their peak on slightly different curves. Second, a short, intense window tempts you to pour budget in fast, which is exactly when cost per sale tends to spike. We needed to scale hard and stay honest about efficiency at the same time.
We treated Mother's Day as a controlled ramp, not a firehose. Budgets moved up geo by geo, each with its own ceiling, and we leaned on the formats that scale cleanly instead of the ones that only look good at small spend.
The US Advantage+ Shopping campaign with a cost cap was the standout — the most reliable way to add spend without watching cost per sale run away. We made it the default for the account.
Australia was efficient at low spend but softened as we pushed it — same audience pool absorbing far more budget. We flagged the ceiling and capped peak spend there instead of forcing it.
We opened TikTok and Reddit alongside Meta and Google to widen the top of funnel. Both behaved like real channels, not experiments — Reddit landed above breakeven in its first season.
The static ads did the scaling work. The strongest was a simple "no typing needed" explainer that walks a buyer through how a spoken interview becomes a real book — it more than doubled its spend back at the platform level and stayed our top performer through the window. We paired proven statics with UGC video from the brand's own creators, so the account always had a fresh angle to rotate in as fatigue set in.
When you scale a peak on purpose, reported ROAS compresses — that's the trade, not a failure. We pushed Meta spend up sharply to chase Mother's Day demand and revenue followed, so the blended platform number dipped during the peak and then settled back above breakeven the moment we eased off. The truth of the season lived in first-party store revenue, not the last-click number any single platform reported.
The headline figures below are the brand's own store revenue, year over year — the hardest number to argue with. We've kept the actual dollar totals out; those belong to StoryKeeper. Platform-reported figures are labelled as such.
Treat the peak as a controlled ramp, not a firehose. We raised budgets geo by geo, each with its own ceiling, and leaned on formats that scale cleanly — an Advantage+ Shopping campaign with a cost cap was the most reliable lever. Store-wide revenue rose 114% YoY, with the US up 250%.
Because you are deliberately buying more demand than the efficient core. When we pushed Meta spend up to chase Mother's Day, revenue followed but reported ROAS compressed during the peak, then settled back above breakeven once we eased off. That is the expected trade — the real signal is first-party store revenue, not a single platform's last-click number.
Attribution. Without a server-side connection, pixel-only tracking misses a chunk of conversions, so Meta under-reported Canadian sales even though the store's own Canadian revenue was up 142% YoY. Installing server-side tracking ahead of the next peak was our top technical recommendation.
We'll show you how to scale the peak without blowing your cost per sale — and which channels are worth opening before it hits. Free audit.
Get a free audit