Why toy and baby stores overspend when they optimise for ROAS

For a toy store, the gap between ROAS and POAS is a timing problem. Much of the year's revenue arrives in a short season, and ROAS looks its best in exactly those weeks. But the peak also brings more competition for clicks, holiday discounting and, after it ends, a wave of gift returns that arrive once the budget is spent.

The figure above sets the scale. Frédérique Tutt, Circana's global toys industry advisor, said in an August 2025 Toy Association release that "the second half of the year accounts for over 60 percent of annual toy sales". Mattel's annual report describes the business the same way: "highly seasonal, with consumers making a large percentage of all toy purchases during the traditional holiday season".

How the gap forms

Peak demand at peak click prices

Demand and competition rise together. More advertisers bid for the same shoppers before the holidays, and click costs can rise with them. ROAS can hold up, because order values rise too, while the profit left after ads shrinks.

Discounts the peak doesn't need

Holiday promotions are designed to win shoppers who are comparing gifts. Some of those shoppers were going to buy anyway. A discount at the peak lowers margin on every order it touches, including the ones it didn't create, and ROAS reports the discounted revenue as a success.

Gift returns after the season

Retailers told the NRF and Happy Returns they expect 17% of holiday sales to be returned, across all categories. Those returns land in January. The December ROAS never changes to reflect them.

Baby products follow a different pattern

Baby essentials are bought steadily through the year and again and again as a child grows. Their gap looks more like any replenishment category: first-order ROAS understates them if customers reorder. Keep them out of the seasonal toy picture, in their own campaigns, so the bidding doesn't learn December's behaviour for products that sell all year.

Working out your own gap

• Measure POAS by season, not by year.
• Hold December's numbers open until February, so gift returns are counted against the season that caused them.
• Compare discounted and full-price orders from the peak, to see what the promotions actually added.
• Tag gift orders. If your checkout captures gift messages or gift wrap, tag those orders and compare their return rate with everyone else's. That tells you how much of January's return wave is gifts, and how much a gift-led campaign really keeps.

How to close it

1. Set peak-season targets from last year's return-adjusted POAS, not from ROAS.
2. Limit peak discounts to the products and audiences where they change behaviour.
3. Split baby essentials from toys in campaigns and reporting.