Every Q4, the same story repeats. CPMs spike on Google and Meta, auctions get crowded, and the shopper you could acquire cheaply in September costs three times as much by Black Friday. Mobile e-commerce user acquisition doesn’t have to follow that script.
How? The fix is to diversify where you find shoppers before the rush starts, not after.
Global online holiday sales hit $1.29 trillion in 2025, up 7% year over year, and mobile drove 56.4% of that revenue for the first time. That’s not a niche shift. It’s the main event, happening on phones. Mobile Original Equipment Manufacturers (OEMs) like Samsung, Xiaomi, and OPPO sit at the center of it, since they control the device before any app store algorithm or ad auction gets a say.
Here are five plays AVOW has run with e-commerce clients that turn Q4 pressure into incremental growth, not simply higher CPIs.
TL;DR
- Mobile e-commerce user acquisition needs to start in August and September, not November.
- Mobile OEM inventory covers 86% of the global Android market and 1.85 billion daily active users, largely untapped by Google, Meta, and Apple auctions.
- Dynamic Preloads let you be the first app a shopper sees on a new device, before Black Friday competition peaks.
- Down-funnel metrics (basket size, purchase rate, ROAS) matter more than raw install counts once you’re inside peak season.
- Retargeting doesn’t end on Cyber Monday. December re-engagement is where seasonal shoppers become repeat customers.
1. Start Diversifying Before October, Not After
The most successful e-commerce marketers don’t wait for Black Friday to test new channels. They test in August, when CPMs are still low, and there’s room to experiment without burning through the budget on a live event.
This matters because mobile OEM inventory operates almost entirely outside the Google-Meta-Apple auction. AVOW’s OEM partnerships cover 86% of the global Android market and reach 1.85 billion daily active users across Samsung, Xiaomi, HUAWEI, Vivo, OPPO, OnePlus, realme, HONOR, and more. For an e-commerce app already maxing out its usual channels, that’s untapped supply, not a smaller slice of the same audience.
Milanuncios felt this pressure directly. Rising competition from Temu, Shein, and Miravia drove up acquisition costs, and ad fatigue eroded performance. Running Xiaomi and OPPO placements alongside its existing mix delivered 30%+ quarter-over-quarter user growth and a 23% reduction in eCPA, without touching the channels it was already saturating.
2. Own the Unboxing Moment With Dynamic Preloads
By October, shoppers are already researching what they’ll buy in November. Apps that show up first on the device itself have an advantage no banner ad can match.
Dynamic Preloads (also called Play Auto Install or Out-of-Box Experience) lets your app appear as a recommended download the moment a user sets up a new phone. The user opts in, which means every resulting install is an engaged one. There’s no SDK integration required, as it runs on a cost-per-install model, and campaigns can launch within about two weeks of a commercial agreement, fast enough to catch the October-to-November window.
Magalu, one of Brazil’s largest e-commerce platforms, used Dynamic Preloads across multiple OEMs alongside Xiaomi GetApps featuring (app featuring on alternative app stores) during peak-seasonality moments. The result was a 9x increase in year-over-year install growth and a 50% scale-up in OEM media investment, run in parallel with existing programmatic, search, and social spend rather than replacing it.
3. Target Purchase Intent, Not Just Install Volume
Raw installs are the easiest metric to chase and the least useful one during shopping season. What actually matters is whether the person who installs is likely to buy.
On-device targeting makes this possible in ways in-app networks can’t. Appographic targeting, which reaches users based on real app usage patterns, lets you find people who already use apps that signal shopping intent, rather than guessing from demographics alone. Joom used this kind of targeted media buying, paired with exclusive Xiaomi placements across Europe and CIS markets, to activate 10 new geos while keeping performance tied to actual purchases rather than downloads.
The Results Speak in Purchases, Not Just Installs
Joom’s campaign generated 60,000 monthly installs, 15,000 purchases, and 150% ROAS. Installs were the input. Purchases were the point.
4. Optimize for Basket Size and ROAS, Not Just CPI
Once a shopper installs, the job isn’t done. Shopping season success shows up in average order value and return on ad spend, so your optimization targets need to shift accordingly.
“AVOW continuously introduces new ideas and innovative solutions for OEM inventories, with an approach that keeps us ahead of the curve in our advertising efforts. Their ability to identify and leverage opportunities in OEMs has been instrumental in diversifying our user acquisition investments and connecting with our target audience effectively.”— Marina, Diogo e Gabriela, Marketing Team, Magalu
Magalu’s team tested a range of OEM ad products, from display and interstitials to app store placements, and tied business models to CPI, CPC, CPM, or CPA, depending on which metric each placement performed best against. That flexibility produced a 25%+ increase in average basket size alongside a 12% banner click-through rate, proof that the right optimization target changes what “winning” looks like.
5. Keep Retargeting After Black Friday Ends
Singles’ Day on November 11, Black Friday on November 27, and Cyber Monday on November 30 will absorb most of your team’s attention this year. That’s understandable, but it’s also where many e-commerce apps stop planning.
Cart abandoners get the obvious retargeting budget, but they’re only part of the audience. Lapsed buyers, high-value browsers, and category viewers each convert differently and deserve their own messaging, not one generic “come back” push. Treat Black Friday and Cyber Monday as full-funnel moments by warming shoppers beforehand, converting them during the event, and re-engaging them after.
December is where that re-engagement pays off. Boxing Day, year-end sales, and New Year’s promotions drive a second wave of activity, and shoppers who bought once during peak season are far cheaper to bring back than new users are to acquire from scratch.
Shopping Season Rewards Whoever Prepares First
None of these five plays require you to replace your existing marketing mix. They require adding a channel that most of your competitors haven’t tested yet, and starting before everyone else scrambles.
You don’t win the shopping season in the last two weeks of November. It’s won in August, while CPMs are low and OEM campaigns are still cheap to test.
See how mobile OEM advertising fits your Q4 roadmap. Book a demo
About the Author
