231% remarketing conversion uplift on US Android, as the activity lifts Android purchase rates 50% more than iOS globally
40% surge in Germany’s consumer spend share on Android as India loses ground despite leading install growth
2x increase in iOS install fraud in France and the UK even as global fraud dropped 34%
Re-engagement dominance hits a new peak as acquisition thins out
The eCommerce marketer’s playbook has run on the same assumption for years: growth means new users. Spend more on acquisition, widen the top of the funnel, convert a percentage into buyers, repeat. That model is changing.
Across the markets and platforms analyzed in this report, the top of the funnel is narrowing. Paid installs are declining broadly, not because demand for eCommerce apps is weakening, but because the cost and complexity of acquiring new users has risen past the point where acquisition-first strategies deliver consistent returns. Competitive saturation in mature markets makes every incremental install more expensive. And geopolitical disruption, most notably the tariff-driven impact on China-based advertisers’ activity in the US, reshuffled billions in ad spend across regions in ways that no annual plan anticipated.
The market’s response has been decisive and nearly universal: shift spend from finding new users to re-engaging existing ones. Remarketing now dominates eCommerce ad budgets on both platforms by a wide margin, and the conversion data validates the bet. In market after market, re-engagement campaigns are delivering value. This transition appears structural, not a temporary reaction to a difficult year.
That shift creates new questions. If the growth engine now depends on extracting more value from existing user bases, which markets are converting engagement into revenue, and which are generating volume without commercial return? The answer varies heavily by geography and platform, and the gap between install growth and consumer spend is one of the most important dynamics in this dataset.
Meanwhile, the infrastructure underneath is under pressure. Fraud is migrating toward the highest-value traffic. Attribution complexity is growing as re-engagement mechanics create new measurement friction. AI-powered analytics are embedded in the daily workflow, but most teams are using them to retrieve answers rather than diagnose problems. For marketers planning the 2026 holiday season, that gap matters: a re-engagement-led strategy is only as good as the decisions behind it.
This report covers October 2024 through March 2026, analyzing ad spend, conversion trends, consumer revenue, buyer behavior, web-to-app patterns, fraud, and AI analytics usage across 13+ markets on both major mobile platforms.
“The journey from discovery to purchase to loyalty is centered on the best surface to move a consumer forward, given the signals available. That means using channels like CTV, social, and creator content to generate demand, with clear and seamless paths into web or app.”
China’s iOS cooldown reshapes UA spend as India claims Android in 2025
For marketers planning the 2026 holiday season, the Q4 comparison shows that globally, iOS user acquisition ad spend posted a slight decline of 5% from Q4 2024 to Q4 2025, while Android posted a slight increase of 5%.
The US tells a different story: After a massive budget spike in the US last year from China-based retailers, overall iOS investment fell 57% year-over-year in Q4 and Android dropped 23%, both reflecting a tariff-driven impact on China-based advertisers’ activity that began mid-year. The contrast with other markets is striking. India’s Android UA grew 63% Q4-on-Q4, which was the strongest gain of any market in the dataset as India-based advertisers filled the space vacated by China-based platforms.
Brazil’s iOS grew 36% over the same period, suggesting continued iOS UA momentum in Latin America heading into the 2026 holiday season. France was the weakest European market on iOS, down 31% Q4-on-Q4, while Germany and the UK held closer to the global average at -11%.
The Q1 comparison reinforces the trend. Globally, iOS fell 51% from Q1 2025 to Q1 2026 and Android dropped 17%. In the US, iOS dropped 69% Q1-on-Q1 as China-based platforms continued to reduce iOS budgets in the US in early 2026. For context on what drove the mid-year spike in 2025: tariff pressure pushed China-based apps to reallocate some budgets from the US toward Western Europe, with iOS UA in Germany jumping 170% YoY and France more than doubling in Jan-May 2025, per our State of eCommerce 2025 report. The Apr-May 2025 global iOS peak was a consequence of that rotation, not a sign of broad-based iOS momentum.
Meanwhile, India-based brands overtook Chinese brands as the dominant source of Android UA spend, with India’s share growing 60% while China’s fell 25%. On iOS, China’s share declined 28%, with the US HQ share growing 74% and the UK’s more than doubling. Despite that decline, Chinese apps still held 53% of iOS UA spend in the second period (Oct 25-Mar 26), meaning the competitive dynamic on iOS has shifted but not fundamentally changed.
It is worth noting that while China-based brands are running aggressive UA campaigns across global markets, Indian-headquartered apps spend predominantly domestically. India’s growth in Android HQ share therefore reflects domestic demand rather than international expansion.
Platform asymmetry in UA spend is structural and market-specific. The English-speaking Western markets are decisively iOS-led: the US and UK both sit at 84–85% iOS share of UA spend, with Japan close behind at 79%. Continental Europe follows the same pattern but not as significant: France at 69%, Germany at 67%. The picture inverts sharply in emerging markets: India allocates 89% of UA spend to Android, Brazil 65%, Mexico 62%.
User acquisition ad spend trend (normalized)
UA ad spend split by platform
UA ad spend split among top markets
UA ad spend split by country headquarters
Remarketing spend in Q4 surged 53% on iOS and 37% on Android
Globally, Android remarketing spend surged 37% from Q4 2024 to Q4 2025, and iOS jumped 53%. Both moves are genuinely broad-based: 10 of 13 analyzed markets grew Android remarketing Q4-on-Q4, and 12 of 13 grew iOS. France was the only market to decline on both platforms. The US is the main exception on Android, falling 19% Q4-on-Q4, as iOS remarketing grew 30% in the same period.
India was the standout growth market: Android up 207% Q4-on-Q4 and iOS 287%. The UK grew steadily on both platforms, up 27% Android and 26% iOS. For marketers planning the 2026 holiday season, the Q4 breadth matters: re-engagement budgets grew across most geographies, suggesting advertisers are leaning harder on their existing user base in the holiday shopping rush.
The Q1 picture shows iOS momentum accelerating even further. Globally, iOS remarketing grew 75% Q1-on-Q1, and the growth is widespread: the US grew 58%, Germany 66%, France 47%, Spain 105%, Brazil 162%, Mexico 64%, and South Korea 64%. That is a strong and consistent signal across markets with very different profiles.
Android Q1 tells a different story: the global +38% is driven almost entirely by India (+390%) and Indonesia (+96%), while most other markets declined. The US, Germany, France, Spain, Brazil, and Mexico all saw Android fall while iOS rose simultaneously, suggesting advertisers in those markets are actively shifting re-engagement budgets toward iOS rather than simply cutting spend.
On the HQ side, India-based brands’ Android remarketing share jumped from 17% to 48% between the two periods while China fell from 41% to 19% — a more dramatic shift than on UA. On iOS, China grew its share from 43% to 50%, while the US HQ share fell from 31% to 17%, suggesting China-based apps prioritized re-engaging existing iOS users even as they scaled back on new acquisition.
Platform asymmetry mirrors UA but with iOS dominance even more pronounced. The UK allocates 90% of remarketing spend to iOS, the US 84%, Germany 70%. Emerging markets invert this: India 78% Android, South Korea 79%. Japan splits evenly 50/50, a notable contrast to its iOS skew on UA.
Remarketing ad spend trend (normalized)
Remarketing ad spend split by platform
Remarketing ad spend split among top markets
Remarketing ad spend split by country headquarters
Remarketing dominates ad spend as UA’s share shrinks across both platforms
Android UA share dropped from 22% in Q4 2024 to 17% in Q4 2025, meaning remarketing’s share rose from 78% to 83%. On iOS, UA share fell from 17% to 11%, with remarketing picking up the difference and reaching 89% of total iOS spend in Q4 2025.
The Q1 comparison shows the shift accelerating even further. Android UA share dropped 42% from 26% to 15% Q1-on-Q1, with remarketing climbing to 85%. On iOS the swing is more dramatic: UA dropped from 23% to just 8%, meaning remarketing now accounts for 92% of all iOS ad spend globally in Q1 2026.
For marketers planning the 2026 holiday season, this is a meaningful signal. Re-engagement continues to be the dominant activity by a wide margin on iOS, and is growing rapidly on Android. Advertisers appear to be betting on converting their existing user base rather than growing it heading into the holiday window.
The pattern holds across most markets but with variation. Germany saw iOS remarketing climb to 85% of total iOS spend in Q4 2025, up from 77% a year prior. France is the exception where Android UA share grew 18% Q4-on-Q4, suggesting French advertisers bucked the trend and held acquisition investment on Android. The US also grew Android UA share 10% Q4-on-Q4 while iOS remarketing continued to dominate at 89%.
Share of ad spend by marketing activity
Paid installs declined broadly while remarketing conversions surged on iOS across all listed markets
On iOS and Android, the same paid install declines produced completely opposite remarketing conversion outcomes. Across both charts together, a more specific pattern emerges: on iOS, paid installs fell in 16 of 21 markets, yet in every single one of those 16, remarketing conversions grew simultaneously. Installs contracted while re-engagement converted better.
This is not a market-by-market anomaly. It held without exception across markets as different as the US (-37% installs, +15% remarketing), Germany (-48% installs, +53% remarketing), Spain (-58% installs, +73% remarketing), and the UAE (-73% installs, +126% remarketing). The implication for 2026 planning is direct: on iOS, acquisition became harder and more expensive while re-engagement delivered, and advertisers appear to have responded accordingly.
Android tells a geographically divided story that the individual charts obscure. Nine major markets: the US, Germany, France, Spain, Italy, Australia, Brazil, Mexico, and the Philippines all saw both paid installs and remarketing conversions decline simultaneously. These are markets where the entire performance marketing funnel contracted, not just acquisition.
By contrast, eight other markets saw installs fall but remarketing conversions grow: Ukraine (+86% remarketing despite -6% installs), Indonesia (+62%/-26%), and the UAE (+46%/-47%) — suggesting that re-engagement budgets held and converted even as acquisition pulled back. The two groups are geographically coherent: double contraction is concentrated in Western markets, while re-engagement resilience is concentrated in emerging and Middle Eastern ones.
India is the only major market to grow on all four dimensions: Android paid installs (+66%), Android remarketing conversions (+119%), iOS paid installs (+71%), and iOS remarketing conversions (+75%), making it the sole unambiguous growth market in the dataset for this period.
Year-over-year % change in conversions (Oct 25-Mar 26 vs. Oct 24-Mar 25)
Germany’s consumer spend share surged on both platforms as India and Saudi Arabia lose ground
Consumer spend share shifts between the two periods (Oct 24-Mar 25 vs Oct 25-Mar 26) reveal a different set of winners and losers than the ad spend data. Germany was the biggest gainer on both platforms, where Android consumer spend share grew 40% (6.7% to 9.4%) and iOS 22% (7.8% to 9.5%), making it the most consistently improved market in the dataset.
The UK also grew on both platforms, up 14% on Android and 17% on iOS, overtaking Japan on iOS to become the third-largest iOS consumer spend market in the second period. Both are high-income markets where shoppers appear to be spending more within apps, possibly reflecting a broader shift toward in-app purchasing as a preferred commerce channel in Western Europe.
India’s consumer spend performance runs counter to its install and remarketing trajectory. Despite leading on paid install growth and remarketing conversion volume, India lost share on both platforms, dropping 14% on Android and 19% on iOS. The pattern suggests that India’s install growth is bringing in a large volume of users who are not converting to paying customers at the same rate, or that per-user spend is diluting as the base expands rapidly. Saudi Arabia showed a similar dynamic, dropping 35% on Android and 20% on iOS — a notable contraction for a market that typically indexes high on consumer willingness to pay.
Brazil held its position as the top Android consumer spend market in both periods but declined 11%, suggesting its lead is softening. On iOS, the US share was completely unchanged at 15.2% across both periods: the only market to hold its share exactly flat, which points to a stable, deeply entrenched spending base that neither grew nor contracted relative to the rest.
Share of in-app purchase revenue by country
South Korea leads buyer loyalty on Android with Brazil showing the sharpest platform divide
South Korea’s dominance on Android buyer loyalty with a 7.53% loyal rate against a dataset average closer to 4% is almost double the next highest market (Saudi Arabia at 5.09%). That gap suggests a structural difference in how Korean ecommerce apps monetize, likely driven by the dominance of highly gamified loyalty mechanics and subscription-first app design that are embedded in the Korean app ecosystem. The UK leads Android regular buyers at 11.21% despite having a smaller absolute user base than markets like India or Brazil, which points to higher monetization efficiency per user rather than volume. On iOS, South Africa tops loyal buyers at 10.26%, which is 2.2x higher than its own Android figure, implying that iOS users in South Africa represent a disproportionately high-value segment relative to the broader smartphone population in that market.
Brazil shows the sharpest platform divide on regular buyers: 13.21% on iOS versus 6.62% on Android, an almost 2x difference. This is consistent with iOS users in Brazil representing a meaningfully higher-spending segment than the broader Android base, which dominates by volume in that market. South Africa comes second at 19.49% iOS versus 14.67% Android, though the gap is narrower. South Korea comes closest to parity, with 20.23% iOS regular buyers versus 16.79% on Android — the tightest spread among high-performing markets, suggesting more uniform purchase behavior across both platforms.
The US and Brazil sit at opposite ends of the market size spectrum but share the same characteristic: both rank at the bottom for Android loyal buyer rates (2.83% and 1.61% respectively). For the US this may reflect the sheer scale of the user base diluting loyalty rates; It may also reflect the sheer volume of app choice available to US consumers, where regular buyers face more alternatives and are less likely to consolidate spend with a single app.for
Brazil it points to lower average willingness to pay in a price-sensitive market. On iOS, Brazil’s loyal buyer rate more than doubles to 3.27% — further reinforcing that the platform split in Brazil is not just about volume but about fundamentally different buyer profiles.
Share of buyers by type *
AI Usage Data: Retargeting Drives 27% of Attribution Questions
An analysis of over 8,000 English-language queries from the AppsFlyer AI assistant by nearly 400 eCommerce brands shows that performance data and campaign analysis is the dominant use case, accounting for 1 in 4 classified questions, more than double any other topic. Attribution and measurement (9%), deep linking and QR codes (8%), and technical or data access questions (8%) round out the top categories, with platform navigation and support trailing at 7%.
Breaking down the largest category reveals a usage spectrum. Nearly half (48%) of performance queries are quick-answer requests: preset insight prompts and simple data retrieval. Channel and source analysis makes up 15%, with trend comparisons and campaign-level deep dives adding another 19%. Strategic and diagnostic queries, including anomaly detection, causal analysis, and optimization recommendations, represent just 4%. It appears that most users engage the assistant for immediate, single-dimension questions rather than multi-layered analysis.
Within attribution, questions cluster around two specific areas. Retargeting and re-engagement mechanics account for 27% of all attribution queries, making it the single largest attribution topic. Marketers are trying to understand where retargeting events surface in reports, and how UA and re-engagement views relate to each other. SKAN and iOS privacy (ATT consent, conversion values, postback configuration) represent 16%. Together, these two areas drive 43% of attribution questions, suggesting that the complexity introduced by re-engagement logic and iOS privacy changes creates the most measurement friction for ecommerce marketers.
Looking at where marketers focus along the funnel, campaign-level performance across both UA and remarketing dominates at 64% of funnel-relevant queries. Monetization questions (ROAS, revenue, LTV, purchase analysis) represent 27%, while engagement and retention (sessions, DAU/MAU, churn) account for just 9%. This gap likely reflects how ecommerce marketers prioritize their time: campaign optimization is the daily operating rhythm, while retention analysis appears to be either deprioritized or handled through other tools.
On question sophistication, 78% of queries fall in an intermediate band: users who specify what they want but typically frame questions around a single metric or dimension. Roughly 12% demonstrate advanced analytical framing with multi-variable breakdowns, causal reasoning, or decision-oriented asks. The remaining 11% are basic navigation or preset prompt clicks. This distribution suggests that most users have solid working knowledge of their analytics platform but are not yet leveraging the AI assistant for the kind of layered, diagnostic analysis it is capable of supporting.
AI Assistant questions split by type among eCommerce marketers *
Black Friday drives iOS web-to-app conversions to 2x baseline as India surges and Brazil fades
Web-to-app — when a brand directing its own website visitors to install its app — is showing a clear seasonal trend. In ecommerce, this channel matters because app users typically have higher purchase frequency and lifetime value than web users, and converting a web visitor into an app user is one of the most efficient LTV moves available.
A web-to-app play also takes into account scenarios in which users are more likely to be acquired on the web as they don’t need to an install an app which is the acquisition event in that platform. In other words, web-to-app offers the best of both worlds in an increasingly omni-channel digital reality.
The November signal is the most visible pattern in the dataset. Global iOS web-to-app conversions surged 66% to correspond directly with Black Friday and Cyber Monday. In the US, iOS November conversions reached almost 2x the baseline rate for the rest of the year. The UK followed a similar pattern.
Critically, this November spike is almost entirely an iOS story: global Android November 2025 conversions came only 11% above its baseline, representing a fraction of the iOS uplift. This suggests that ecommerce brands running web-to-app campaigns during the holiday window are investing heavily in iOS re-engagement, where app users convert at higher rates and are worth more to acquire.
India tells a different story. Android web-to-app conversions climbed steadily with the H2 2025 average of more than double the H1 2025 average, reflecting rapid scaling of web-to-app infrastructure among India-based ecommerce players as they invest in deepening their app user base.
Brazil moved in the opposite direction. November 2024 was Brazil’s peak but both platforms declined sharply from there. By Q1 2026, iOS averaged well below where Brazil started. It appears that Brazilian ecommerce brands pulled back on web-to-app investment after an initial push, possibly reflecting budget reallocation or lower-than-expected conversion returns on the channel.
Web-to-app installs trend by platform (normalized)
Android users need more convincing as remarketing lifts CR by 50% more than iOS
Remarketing improves install to purchase conversion rates everywhere, which is expected. What this data reveals is just how much it improves them, and where the platform gap is widest. Globally, remarketed campaigns convert users into purchasers at a rate 118% higher on iOS and 177% higher on Android than campaigns without remarketing. The Android advantage is consistent across almost every market, but the size of that advantage varies dramatically and points to fundamentally different user dynamics by platform.
The markets with the largest Android-iOS divergence are the most instructive. In Brazil, remarketing lifts Android conversion by 211% but iOS by only 69%. South Korea shows a similar pattern at 219% Android versus 101% iOS. The most likely explanation is that iOS users in these markets are already self-selecting. They represent a smaller, higher-intent segment of the smartphone population, so remarketing adds less incremental lift because baseline purchase intent is already elevated. Android user pools in these markets are much larger and more heterogeneous, meaning more users need that additional touchpoint to convert.
The US is notable for a different reason: it has the highest absolute Android uplift in the dataset at 231%, despite being a mature market where iOS dominates ecommerce. That suggests Android users in the US are significantly harder to convert without re-engagement than their iOS counterparts, and that remarketing is doing heavy lifting to bridge that gap.
Indonesia and Mexico are the two exceptions to the platform divergence pattern. Indonesia sees near-identical uplift on both platforms — 181% iOS versus 183% Android — and Mexico is similarly balanced at 91% and 95%. Germany is also close at 168% iOS versus 180% Android. In these markets, the purchase intent profile of iOS and Android users appears more similar, making remarketing roughly equally valuable regardless of platform.
Remarketing uplift in the install to purchase conversion rate *
Install fraud is falling in most markets but rising sharply on iOS in France and the UK
Globally, install fraud declined from 12.4% to 10.3% on Android and from 16.6% to 11.0% on iOS between Q1 2025 and Q1 2026, marking a meaningful improvement on both platforms. But the global trend masks a sharp divergence at the market level. Indonesia, South Africa, and Saudi Arabia all saw Android fraud rates more than halve over the period, suggesting that improved detection and cleaner traffic sources are taking hold in markets that were previously more exposed. India iOS fraud fell from 13.1% to just 2.4% — the sharpest decline in the dataset.
The countertrend is concentrated in Western Europe on iOS. France iOS fraud climbed from 11.8% in Q1 2025 to 23.0% in Q1 2026, a near doubling with no sign of reversal as every quarter was higher than the last. UK iOS fraud also surged, from 11.4% to 20.0%. These markets are seeing sustained growth in fraudulent iOS installs even as the global average improves, which may reflect fraudsters shifting focus toward higher-value iOS traffic in mature markets where CPIs are elevated and the payoff per fraudulent install is larger.
On the media type chart, affiliates stand out with both the highest fraud rate and the worst trajectory. The Q4 2025 peak of 41.2% aligns directly with peak holiday UA season — shopping advertisers surge affiliate spend for Black Friday and Christmas, onboarding new networks with less history and looser controls, creating the optimal fraud window. The partial Q1 2026 recovery to 35.6% confirms a seasonal component, but the underlying baseline rose from 27% to 35%+ even in non-peak quarters, pointing to a structural deterioration.
Organic fraud tells the opposite story directionally — declining from 15.5% to 10.7% — but the implications are more damaging than the numbers suggest. Organic is every company’s internal performance benchmark, so when it is polluted by fraudsters washing fake installs as self-discovered traffic, it distorts the baseline that every other channel gets measured against. At the other extreme, self-reporting networks (Google, Meta, TikTok, Apple, Snap) register just 0.9% fraud across the period — their closed-loop measurement leaves far less room to manipulate attribution.
App install fraud rate by platform
App install fraud rate by media type
- The data in this report shows remarketing now accounts for a vast majority of iOS eCommerce ad spend for apps. How does that shift from acquisition toward re-engagement play out across channels beyond mobile apps — especially web, but also CTV or retail media — and where do you see the biggest gaps in how marketers connect those touchpoints?
-
On iOS, the shift toward remarketing reflects a broader change in how performance budgets are being allocated. Marketers are shifting from broad, generic reach to approaches that mix proven intent with finding new shoppers. Those same marketers are leaning into their first-party audiences and using them to re-engage users across web, CTV, and retail media, not just within apps.
On the open web, dynamic display is increasingly used to bring high-intent users back to products or categories they have already explored. Within retailer environments, on-site and retail media placements do the same job closer to the point of purchase, reconnecting brands with shoppers who have already shown intent. CTV is becoming an extension of those efforts: it’s no longer only about reach, but about using commerce signals to build audiences and then measuring whether those households come back and purchase, whether on web or in app. CTV is a way to feed the upper size of your funnel pipeline and convert more.
The biggest gap is still identity and governance across these environments. Many marketers are planning and measuring in silos, which makes it difficult to understand whether different touchpoints are reinforcing each other or simply duplicating spend. As those signals become more connected, ideally through a common commerce-data and AI layer, marketers will be better positioned to manage frequency, sequence messaging, and invest behind the paths to purchase that are truly incremental rather than the channels that simply attribute conversions most effectively. - As eCommerce brands prepare for Q4 2026, what does an effective omni-channel strategy look like in practice? Where do most brands underinvest or over-index on a single channel?
-
A strong omnichannel strategy starts with the shopper, not the channel. It maps the journey from discovery to purchase to loyalty and asks a simple question at each step: what is the best surface to move this person forward, given the signals available?
In practice, that means using channels like CTV, social, and creator content to generate demand, with clear and seamless paths into web or app. It also means treating your site and app as media environments rather than just destinations and using those high intent visits to build audiences that can be re-engaged across the open internet and retail media during peak periods like Black Friday and Cyber Week. Platforms that bring those channels together, from planning and activation through measurement, make it easier to execute that strategy consistently.
Where brands fall short is not execution within a channel, but connection across them. Teams often over index on familiar environments, while underinvesting in linking those touchpoints together. As a result, strong performance in one area is not always translating into better overall outcomes.
More advanced teams are starting to shift their approach by setting clearer incrementality guardrails, using a mix of experiments, scenario planning, and more dynamic pacing. They reallocate budget more frequently during peak periods based on what’s actually working, rather than sticking to fixed channel plans. - AI is increasingly embedded in campaign optimization and audience targeting. Where do you see AI making the biggest impact on cross-channel retargeting today, and where is it still falling short?
-
AI is already making a meaningful impact in several areas of cross channel retargeting. One is signal enrichment, where models can infer valuable audience insights from incomplete or delayed data and update those audiences in near real time. This becomes especially important when working across web, app, CTV, and retail media, and hopefully LLMs soon.
Another area is budget allocation. Rather than optimizing channels in isolation, AI can help distribute spend, which is particularly useful during periods such as Q4 when conditions change quickly.
Creative is also evolving. Generative tools are making it easier to produce and test tailored assets across different audiences and environments, helping brands maintain a cohesive story while adapting to each surface.
There are still limitations to address. Data fragmentation can limit effectiveness, since disconnected data and measurement frameworks make it harder to understand impact across the full journey. In practice, AI tends to perform best where consented data is reasonably well connected, governance is clear, and objectives, including longer-term value, are clearly defined. - What is the role of measurement in engaging shoppers across the entire journey and making every commerce moment smarter?
-
A lot of people are talking about “incrementality.” The reality is, we’re unlikely to ever land on a universal, clean definition of it in our industry. It’s inherently imperfect. But that doesn’t mean it’s out of reach. The real opportunity is to get as close as possible to it through better measurement. And that’s the point: measurement isn’t a proxy for incrementality, it’s the only practical way to approach it.
Measurement is what connects modern commerce strategies. Without it, omnichannel activity becomes a collection of disconnected bets and with it, every interaction can inform the next decision. It plays a central role in building confidence in decision making. When teams shift budget across channels, they need a consistent way to understand whether those changes are driving incremental outcomes, not simply redistributing attribution.
Measurement also helps create a more coherent customer experience. When brands understand how and where someone has engaged, they can manage frequency, reduce redundancy, and guide audiences more naturally from discovery through purchase and into loyalty.
Finally, it underpins responsible use of AI. Models are only as effective as the signals and objectives they are given. If optimization is based only on short term revenue, outcomes will skew toward aggressive tactics. Incorporating metrics like margin, retention, and lifetime value leads to more balanced decisions.
As the industry evolves, the focus is shifting from evaluating individual channels to understanding how sequences of touchpoints influence specific shoppers, and how AI can orchestrate those sequences more intelligently. That shift, powered by robust measurement and commerce data, is what allows brands to make more informed decisions and drive more durable growth.
- The report shows remarketing now dominates eCommerce ad spend globally, but the speed of that shift varies by market. In your region, how are eCommerce brands actively planning around a re-engagement-first strategy, or is acquisition still the default mindset heading into Q4?
-
In North America, acquisition is still the default motion but the smarter brands are quietly shifting budget toward re-engagement as iOS UA costs climb and returns compress.
The brands winning heading into Q4 aren’t choosing between the two. They’re using UA to feed the remarketing pool, then letting retention do the heavy lifting.
What’s driving this? In Retail and QSR especially, we’re seeing brands go deeper with their existing customers: new incentives, more touchpoints, stronger relationship-building across the full lifecycle. The result: more repeat purchases, higher LTV, and a customer base that actually shows up when it matters most.
It’s a smarter use of the data they already have.
Q4 is closer than it looks. The brands that arrive at holiday with a segmented, multi-signal approach not just a media blitz will be the ones converting intent into revenue when spend peaks. UA fills the funnel. Retention closes it.
The brands that treat those as separate strategies will feel it. - iOS and Android play very different roles depending on the market. How do the eCommerce brands you work with in your region think about platform allocation — do they treat iOS and Android as separate strategies with different goals, or is it still one budget split by volume?
-
Most NA brands still treat platform allocation as a single budget split by volume. It’s leaving money on the table.
iOS skews higher-value and drives disproportionate LTV: that part is well understood. What’s less understood is Android. It carries the volume, and the data shows remarketing lifts Android conversion significantly. Yet it remains chronically underfunded on re-engagement.
Some verticals are already making the shift. QSR brands in particular are leaning into Android as a primary re-engagement surface and the results are backing it up.
AppsFlyer data points to Android re-engagement as one of the highest conversion lift opportunities we’ve seen globally. And North America may be the most underleveraged market for it.
Here’s why: US Android users are a broader, less self-selecting audience. They haven’t opted into high-spend behavior the way iOS users have. They’re not unreachable, they just need that extra touchpoint. The intent is there. The nudge is missing.
For brands heading into Q4 with retargeting budget to deploy, Android re-engagement isn’t a secondary line item. It’s an underpriced edge. - Web-to-app is emerging as a meaningful conversion channel for eCommerce, especially around peak shopping events. Are the brands in your region actively investing in moving web visitors into their apps, or is that still an untapped opportunity for most?
-
The brands that win this holiday season will push web to app. That’s not a prediction it’s already showing up in the data.
Black Friday is becoming an app moment. When you look at the numbers, app isn’t just a channel preference, it’s the destination that matters most during peak season. An app user isn’t just a purchaser. They have a relationship with the brand. They’re easier to reach, easier to re-engage, and they convert more over time.
Yet heading into the biggest spending period of the year, this remains one of the most underleveraged opportunities in North America.
NA brands are sitting on massive web audiences built through paid search and social. But most haven’t built the infrastructure to convert that traffic into app users smart banners, deep linking, frictionless handoffs from browser to app. The mechanics aren’t complicated. The delta in purchase frequency and LTV on the other side is measurable and significant.
Web gets the click. App builds the customer.
Brands that close that loop before Q4 peaks won’t just see better holiday numbers, they’ll enter 2027 with a higher-value user base than the brands that didn’t. - The data shows fraud is declining in some regions but rising sharply in others, particularly on iOS in Western Europe. What is the fraud conversation like with eCommerce brands in your region — is it a top priority, an afterthought, or somewhere in between?
-
Fraud is declining and that’s partly a success story. Brands are getting smarter, teams are more connected, and measurement is closing the gaps that bad actors used to exploit.
But fraud is rarely the lead conversation. And in some ways, that’s the problem.
The bigger issue in North America right now isn’t classic install fraud. It’s attribution integrity across a fragmented media mix. When brands are running simultaneously across retail media networks, social, affiliate, and owned app: the gray zones multiply fast. Credit gets misassigned. Budgets follow bad signals. And the damage is quiet, which makes it harder to catch and easier to ignore.
This is the fraud conversation NA brands should be having: not just “are we protected” but “do we actually trust our attribution data end to end?”
In a media environment this complex, integrity isn’t a security function it’s a growth function. - Cross-channel and omni-channel measurement keeps coming up as a challenge for eCommerce marketers globally. In your region, how sophisticated are brands at connecting their mobile app data with web, retail, and other touchpoints — and what is holding the laggards back?
-
Cross-channel and omnichannel measurement aren’t nice-to-haves. They’re the foundation everything else is built on.
The ambition is there. The infrastructure, for most NA brands, isn’t. Not yet.
Most are still operating with siloed tech stacks: web analytics in one place, app attribution in another, CRM somewhere else entirely. The data exists. The connections don’t. And when the web team, app team, and retail media team each own their own P&L, unified measurement stops being a technical problem and becomes an organizational one.
We see that’s the real blocker. Not the tools, the structure.
What we hear consistently from NA brands is that omnichannel measurement is a stated priority at the leadership level, but it stalls in execution because no single team owns the full picture. Everyone is optimizing their slice. Nobody is optimizing the whole.
This is where AppsFlyer can move the needle, not just as a measurement platform, but as the connective layer that makes a unified view of the customer actually achievable. The brands that close this gap before peak season won’t just measure better. They’ll make faster, smarter decisions when it matters most. - AI-powered analytics tools are now part of the daily workflow for many eCommerce teams, but most queries we analyzed are basic data retrieval rather than strategic diagnosis. Does that match what you see in your region, and which brands are using AI more ambitiously?
-
The honest answer? Most NA teams are still in “AI for productivity” mode faster reporting, better copy, smarter segmentation. Useful. But not transformative.
The brands doing it differently are using AI to compress the time between signal and action. Not to make dashboards prettier to connect behavioral data to real-time campaign decisions.
The smartest teams are going a level deeper. They’re leveraging AI not just for smoother workflows, but for the kind of diagnostic thinking that actually moves strategy: not what happened, but why it happened and what to do about it. That’s where AI stops being a productivity tool and starts being a competitive advantage.
There’s another benefit that doesn’t get talked about enough: time. Analysis that used to take teams days can now be done in minutes. That’s not an incremental efficiency gain — it’s a structural shift in how fast a team can learn, adapt, and optimize across a full campaign cycle.
The brands that will pull ahead aren’t using AI to do the same things faster. They’re using it to ask better questions, act on answers sooner, and apply those insights across their holistic marketing strategy in ways that compound over time. - What is the impact of China-based apps in your region? What do you hear from marketers?
-
China-based apps reshuffled the US market and our data shows it clearly. When these brands lpulled back US investment in response to tariffs, it created a real opening for the competitive set. The brands that saw the window moved fast: re-evaluating their media mix, reallocating spend, and stepping into share that had quietly become available.
That moment is a case study in something bigger. Market disruption is a constant. The question was never whether it would happen, it’s whether you’re positioned to move when the window opens.
The longer-term signal is harder to reverse. Chinese brands have permanently reset consumer price expectations in North America. That’s not unwinding. And the response from most marketers isn’t panic, it’s recalibration. Doubling down on retention, loyalty, and owned channels rather than trying to out-acquire platforms with structurally lower cost bases.
That’s the right instinct. You can’t win a race to the bottom on acquisition cost against players built for it. You win by making your existing customers more valuable, more loyal, and harder to poach.
Disruption reveals strategy. The brands that came out ahead weren’t just lucky with timing, they had the measurement infrastructure and the organizational agility to act on what they were seeing in real time. - As your region's eCommerce brands start planning for the 2026 holiday season, what is the single biggest challenge or opportunity you hear them talk about most — and how does it connect to the trends in this report?
-
Heading into Q4, the biggest challenges I see for brands are twofold — and they’re connected.
The first is siloed teams operating without a full view of omnichannel measurement. Web, app, retail media, CRM. Each team optimizing their slice, none of them optimizing the whole. The fix isn’t just technical. It requires teams to step back from their individual metrics and align around the brand’s overarching goal, then identify how each function supports that in its own unique way. Shared measurement is what makes that possible.
The second is preparation or the lack of it. The biggest opportunity heading into holiday is also the biggest risk: the brands that built their remarketing pools and loyalty infrastructure in Q2 and Q3 will have a significant advantage come November. Those that didn’t will be paying a premium to reach audiences they already owned — or competing on price against platforms they can’t outspend.
This isn’t a Q4 problem. It’s a year-round discipline.
The brands that treat performance data as a continuous signal not a quarterly readout will arrive at holiday with better audiences, sharper segmentation, and more efficient spend. The sooner brands act on what they’re seeing now, the better positioned they’ll be when it matters most.
Q4 doesn’t reward last-minute pivots. It rewards preparation. - What are the top 3 trends in eCommerce in 2026 in your region and how will they play out in the holiday season?
-
The top trends shaping NA eCommerce in 2026 come down to one throughline: the brands building the right foundation now will be the ones winning when it counts.
Omnichannel strategy wins: but only with the measurement to back it. Ambition without infrastructure is just aspiration. The brands pulling ahead are the ones connecting the pipes across web, app, retail media, and CRM. And using that unified view to make faster, smarter decisions.
Remarketing becomes the primary growth engine. 92% of iOS ad spend tells you where the market has already moved. LTV customers are the asset. How brands engage with them in-app drives the highest and most durable ROI — not just this quarter, but across the full customer lifecycle..
App is the center of the customer journey. The brands that win holiday will be the ones that made that decision months earlier. Higher LTV, stronger loyalty, web-to-app conversion, in-store QR. App isn’t a standalone mobile channel. It’s the hub.
None of it works without measurement tying it together. Real-time, cross-channel attribution through an unbiased third-party partner is what separates brands that maximize their holiday investment from those flying blind when spend peaks.
The foundation isn’t a nice-to-have. It’s the whole game.
- The report shows remarketing now dominates eCommerce ad spend globally, but the speed of that shift varies by market. In your region, how are eCommerce brands actively planning around a re-engagement-first strategy, or is acquisition still the default mindset heading into Q4?
-
In EMEA, the shift toward re-engagement is real, but it is not evenly distributed. The more sophisticated brands, particularly in the UK, Germany, and France, are actively building re-engagement-first strategies. They have seen the same thing this report shows globally: on iOS, paid installs declined across most EMEA markets while remarketing conversions grew without exception.
What I see in practice is that the best EMEA eCommerce teams are treating their existing app user base as their most valuable media channel heading into Q4. They are investing in push notification strategies, deep link-driven re-engagement flows, and remarketing campaigns that target lapsed buyers with personalized offers timed around key shopping moments: Black Friday, Singles’ Day, pre-Christmas.
That said, acquisition has not disappeared from the conversation. In markets like Poland and Turkey, where app ecosystems are still maturing, UA remains a significant part of the mix. And even in mature Western European markets, brands still need acquisition to replenish their remarketing pools. The difference is that the smart brands now think of UA as feeding the re-engagement engine rather than being the growth strategy in itself. - iOS and Android play very different roles depending on the market. How do the eCommerce brands you work with in your region think about platform allocation — do they treat iOS and Android as separate strategies with different goals, or is it still one budget split by volume?
-
EMEA is one of the most platform-fragmented regions, and the brands that perform best treat iOS and Android as fundamentally different channels with different objectives, different user profiles, and different economics.
The data in this report reinforces why. The UK allocates 90% of remarketing spend to iOS. Germany sits at 70%. They reflect the reality that iOS users in Western Europe are higher-value, higher-intent shoppers, and the remarketing return on iOS is disproportionately strong. Meanwhile, markets like Turkey and parts of Eastern Europe skew heavily Android, where the user pools are larger.
The best EMEA eCommerce brands I work with run separate strategies by platform. On iOS, the focus is retention, re-engagement, and maximising lifetime value from a smaller but more valuable user base. On Android, the play is often broader: acquisition-led, volume-driven, with remarketing used to convert a wider funnel.
Where the report’s conversion data really lands is on the Android opportunity. Remarketing lifts Android purchase conversion rates by 177% globally versus 118% on iOS. In practice, that means many EMEA brands are leaving money on the table by under-investing in Android re-engagement. They default to treating Android as their acquisition platform and iOS as their monetisation platform, without realising that Android remarketing delivers a larger incremental lift precisely because those users need the additional touchpoint to convert. - Web-to-app is emerging as a meaningful conversion channel for eCommerce, especially around peak shopping events. Are the brands in your region actively investing in moving web visitors into their apps, or is that still an untapped opportunity for most?
-
I see web-to-app as a significantly untapped opportunity, and that represents one of the clearest growth levers available heading into Q4.
The report shows a striking seasonal pattern: global iOS web-to-app conversions surged 66% in November, directly aligned with Black Friday and Cyber Monday. In the UK, iOS November conversions reached nearly double the baseline. It is evidence that web-to-app is one of the most efficient paths to converting high-intent holiday shoppers into app users, where they buy more frequently and deliver higher lifetime value.
What makes web-to-app particularly relevant for EMEA is that European eCommerce brands tend to have strong web presences. Many of the retailers I work with drive significant traffic through web-based paid search, social ads, and organic channels. The brands that are investing in web-to-app journeys are seeing results, particularly with use of smart banners and deep-linking. - Cross-channel and omni-channel measurement keeps coming up as a challenge for eCommerce marketers globally. In your region, how sophisticated are brands at connecting their mobile app data with web, retail, and other touchpoints — and what is holding the laggards back?
-
EMEA has some genuinely advanced omni-channel retailers, brands with physical store presence, apps, and a web presence. The core challenge is not ambition; it is infrastructure. Many EMEA eCommerce brands run different tech stacks for web analytics, app attribution, CRM, and retail point-of-sale. Stitching those together into a unified customer view requires both technical integration and organisational alignment.
The brands getting it right in EMEA tend to share a few characteristics: they have invested in web-to-app measurement infrastructure, they are using deep linking to maintain attribution across touchpoints, and they have teams that sit across both digital and retail. What is holding the laggards back is often organisational as much as technical. The web team, the app team, and the retail team each own their own P&L and their own data. Until that changes, until measurement is genuinely cross-functional, omni-channel will remain an aspiration rather than an operating reality for most EMEA eCommerce brands.
Here is the thing: customers do not know that your marketing teams sit in different departments with different KPIs. They expect to move seamlessly across app, web, and in-store, and they expect brands to know who they are across those touchpoints. The disconnect between how brands are organised internally and how customers actually shop is where the real friction lives.
The brands that do this best have the app at the centre of the customer journey. They build dedicated in-store modes so customers can find wishlist items on the shop floor. They use QR codes in-store to drive app downloads at the point of discovery. They centralise loyalty to the app, so customers scan at checkout to earn and redeem points whether they are shopping online or in person. The app becomes the connective tissue between every channel, and that gives both the customer a seamless experience and the brand a unified view of behaviour across touchpoints. - What is the impact of China-based apps in your region? What do you hear from marketers?
-
Temu, Shein, and increasingly TikTok Shop have fundamentally altered the competitive landscape. These platforms operate with a level of UA aggression that most EMEA-native retailers simply cannot match. The report shows that despite a decline in China-based iOS UA share globally, Chinese apps still held 53% of iOS UA spend in the second half of the reporting period. In EMEA specifically, we saw the effects of China-based budget rotation firsthand: when tariff pressure shifted spend from the US toward Western Europe in mid-2025, iOS UA in Germany jumped 170% and France more than doubled.
What I hear from EMEA marketers is a mixture of frustration and forced innovation. The frustration is real: China-based apps have inflated CPIs across Western European markets, making acquisition more expensive for everyone. They have also trained consumers to expect deep discounts, gamified shopping experiences, and near-instant delivery, which puts pressure on EMEA brands’ margins and value propositions.
But the forced innovation part is where it gets interesting. The smarter EMEA brands are responding not by trying to outspend China-based competitors on acquisition, but by doubling down on what those platforms cannot easily replicate: brand trust, product quality, localised customer service, and omni-channel experiences that connect digital and physical retail. They are also leaning harder into remarketing, owned media, and retention, building deeper relationships with existing customers rather than competing for the same top-of-funnel traffic. They are also diversifying their media mix so that they are not overly reliant on one or two channels.
The honest truth is that China-based apps have permanently raised the bar in EMEA eCommerce. Brands that treat this as a temporary disruption rather than a structural shift are going to lose ground - What are the top 3 trends in eCommerce in 2026 in your region and how will they play out in the holiday season?
-
Loyalty and remarketing converge as the primary growth engine. Remarketing accounts for 92% of iOS ad spend and the conversion data validates the shift across nearly every market. The brands that will win Q4 are the ones with deep, loyal customer bases they can re-engage profitably. Loyalty is the commercial engine: brands that centralise loyalty to their app and use purchase history, preference data, and reward mechanics to personalise re-engagement campaigns are seeing significantly stronger conversion rates. The time to build your remarketing pools and loyalty infrastructure is Q2 and Q3, not November.
2. The retail media opportunity. This is playing out on two fronts. Major EMEA retailers like Zalando, Carrefour, and Boots are building their own retail media networks, monetising first-party shopper data. At the same time, brands are increasingly advertising across these networks to reach high-intent audiences closer to the point of purchase. For Q4, retail media offers something traditional channels struggle to deliver: proximity to buying intent. As CPIs rise across paid social and display, partly driven by China-based competition, retail media gives EMEA brands an alternative route to consumers already in a shopping mindset.
3. Omnichannel as a core strategy. iOS web-to-app conversions surged 66% in November globally, and the UK mirrored that pattern. App users consistently deliver higher purchase frequency and lifetime value. Brands investing in smart banners, deep-linked campaigns, and QR codes in-store are converting existing web and in-store traffic into higher-value app users without paying for a new install through traditional UA. With web traffic peaking around Black Friday, the brands with this infrastructure in place will capture the surge. Those without it will watch high-intent visitors browse and leave.
- The report shows remarketing now dominates eCommerce ad spend globally, but the speed of that shift varies by market. In your region, how are eCommerce brands actively planning around a re-engagement-first strategy, or is acquisition still the default mindset heading into Q4?
-
No one in APAC is abandoning acquisition. But the brands with mature user bases are getting sharper about when to switch from “get new users” to “get more from existing ones.”
The trigger is usually geo expansion. If a brand is expanding into a new country, it is UA-first — there is no user base to re-engage. If they are deepening in a market they already own, remarketing takes over. A player like Coupang in Korea, where purchase frequency runs 24x per year, does not need to convince anyone that retention matters. But Coupang expanding into another market? That is a UA story. Same company, different playbook depending on the market.
Another interesting shift I am seeing is with the mega global players from APAC. These companies have historically relied on their on-platform recommendation engines as their core retention strategy, and for good reason. Those engines are extremely good at knowing what you want to buy next. Paid marketing budgets went almost entirely to UA teams, because the product itself handled re-engagement.
That is starting to change. These players are now proactively seeking paid remarketing partners, something they did not do two or three years ago. Part of it is that competition for new users is getting more expensive, especially across multiple geographies at once. Part of it is that even the best recommendation engine only works on 1st party data. - Web-to-app is emerging as a meaningful conversion channel for eCommerce, especially around peak shopping events. Are the brands in your region actively investing in moving web visitors into their apps, or is that still an untapped opportunity for most?
-
It depends on where the brand started. In APAC, the web-to-app story plays out very differently depending on whether a company was born as an app or born as a website.
India’s largest eCommerce players — Flipkart, Meesho, Zepto, Blinkit — are app-first from day one. Their users discover, buy, and repeat inside the app. Web is not the main acquisition funnel. Social is. WhatsApp shares, Instagram ads, word of mouth — all of it points straight to app install. For these brands, “web-to-app” is not really the conversation. They skipped the web step.
The story is completely different with China-origin DTC brands going global — Halara, PatPat, Cider, and a growing wave behind them. These brands start with a website, usually built on Shopify or a Chinese site-builder like SHOPLAZZA, and drive traffic through TikTok creators and affiliate networks. Halara alone works with over 5,000 creators. That works for a while. But as they scale, they launch an app to improve retention, push notifications, and repeat purchase rates. Now their user journey looks like this: TikTok ad → brand website → app install → repurchase. And the measurement breaks at every handoff.
This is where web-to-app becomes a real pain point, not a ‘nice to have’. - The data shows fraud is declining in some regions but rising sharply in others, particularly on iOS in Western Europe. What is the fraud conversation like with eCommerce brands in your region — is it a top priority, an afterthought, or somewhere in between?
-
The harder problem is what I would call “grey zone” fraud, and it is growing fast for APAC based eCommerce who target both the home APAC market and western EU, for two reasons.
First, the rise of creator and affiliate-driven acquisition. When an eCom brand works with thousands of KOLs and affiliate partners across TikTok, Instagram, and local platforms, attribution gets messy. Did that creator actually drive the install, or did they claim credit for an organic user? The more partners you add, the harder it gets to tell. This is not traditional fraud — no one is spoofing an SDK. But the financial impact is real, and most brands do not have the tools to catch it at scale.
Second, web-plus-app fraud. As more APAC-based eCommerce brands run campaigns across both web and app, fraud detection needs to work across both surfaces. A user who clicks a web ad and installs the app creates a cross-channel journey that fraudsters can exploit at the handoff point. The brands I work with that run both web and app campaigns are now asking for unified anti-fraud coverage — something they did not ask for even a year ago. - Cross-channel and omni-channel measurement keeps coming up as a challenge for eCommerce marketers globally. In your region, how sophisticated are brands at connecting their mobile app data with web, retail, and other touchpoints — and what is holding the laggards back?
-
APAC actually has some of the most advanced offline-to-online or offline for online playbooks in the world. The problem is not that no one knows how to do it. The problem is that most brands cannot get their own teams to agree on who owns the customer.
Take Uniqlo in China. They engineered store-to-app conversion down to every touchpoint — QR codes at entrances, fitting rooms, cashier counters, store-exclusive discount barcodes that require the app to redeem, and company-wide staff training before the app even launched. During one campaign period, the results were clear: 10x foot traffic versus prior year, offline order value 34% higher than online. The store drove people to the app. The app drove people back to the store. Both channels won.
But here is what could hold some of the APAC brands back: the internal org chart. Web marketing, app marketing, and retail operations typically report to different leaders with different KPIs. The store team is measured on foot traffic and in-store revenue. The app team is measured on downloads and in-app purchases. Nobody is measured on “did the customer who walked into the store become a repeat buyer through the app 6 months later?” - AI-powered analytics tools are now part of the daily workflow for many eCommerce teams, but most queries we analyzed are basic data retrieval rather than strategic diagnosis. Does that match what you see in your region, and which brands are using AI more ambitiously?
-
The question assumes AI in eCommerce is about analytics tools. In China, it has moved well past that. Outside of China, most markets are still in the “experimenting” phase. There is a lot of buzz — “we are piloting AI for X,” “we are exploring AI-powered Y” — but very little is live at scale with real transactions flowing through it. The AI sits alongside the business. It helps people look at data. That is useful, but it is not what China is doing.
In China, the leading platforms skipped the experimentation stage and went straight to execution. The clearest example is what Alibaba has done with Qwen, their own large language model. Qwen is not a chatbot bolted onto Taobao. It is embedded directly into the transaction flow. A shopper can search, discover, compare, buy, and pay — all inside a single AI conversation.
The second point worth calling out is that China has built its own LLM moat, quite separate from the Western AI ecosystem. Alibaba has Qwen. ByteDance has Doubao, which reached 226 million monthly active users by early 2026. Tencent has Hunyuan embedded in WeChat commerce. Baidu has Ernie at 202 million MAU. And then there is DeepSeek, now being used by livestream platforms for product copy generation and ad screening.
It is not about whether eCommerce teams use AI for basic data retrieval or strategic diagnosis. In China, the question has already moved on: the AI is the commerce layer itself. - What are the top 3 trends in eCommerce in 2026 in your region and how will they play out in the holiday season?
-
1. AI moves from a productivity tool to core infrastructure: The conversation around AI in eCommerce has been dominated by “we are using AI to do X faster.” That framing is already outdated. In APAC — especially in China — AI is becoming the infrastructure itself.
2. The line between online and offline is disappearing: This has been talked about for years, but what is different in 2026 is that the measurement and technology barriers are actually being removed. Deep linking connects web sessions to app installs. Geofencing closes the loop between push notifications and store visits. Cross-channel attribution is finally making it possible to answer “did this online ad drive that offline purchase?”
3. Brand equity is getting harder to defend: Here is the consumer trend that should worry every premium eCommerce brand heading into the holidays: it is not just one cheaper alternative anymore. There are dozens. Now more and more consumers who buy dupes are high-income, not budget shoppers. This is not a price sensitivity problem. It is a value perception problem. TikTok, SHEIN, and Temu have made discovery frictionless. A consumer like me searching for yoga pants will see ten alternatives before they ever reach the brand’s own page. Every holiday season amplifies this — when promotions are everywhere and comparison is effortless, the brand that cannot clearly justify its price premium loses share to the brands that never had one to protect. For APAC eCommerce brands, the holiday playbook is shifting: it is less about driving traffic and more about defending why your product is worth the price when the customer has already seen five cheaper options.
Remarketing now accounts for the vast majority of eCommerce ad spend on both platforms, and conversion data validates the shift. In markets where paid installs declined on iOS, remarketing conversions grew without exception. For holiday planning, the signal is clear: explore shifting more budget toward re-engaging existing users rather than fighting for increasingly expensive new installs, especially on iOS where acquisition costs continue to rise.
Remarketing lifts Android purchase conversion rates significantly more than iOS across most markets, because Android user pools are larger and more heterogeneous. Markets like Brazil, South Korea, and the US show the widest platform divergence. Explore increased Android remarketing investment in markets where your iOS re-engagement is already mature — the incremental return on Android is likely higher than pushing iOS further.
iOS install fraud is rising sharply in Western European markets even as global fraud declines. Agency-sourced traffic spiked during the last holiday season, when advertisers onboard new networks with less history and looser controls. Review traffic quality by source and market before scaling Q4 budgets, and consider tightening validation on affiliate and agency channels during peak spend periods.
Install growth and consumer spend growth are diverging across key markets. Some of the fastest-growing markets on installs and conversions are losing consumer spend share, while others with modest acquisition metrics are gaining disproportionate revenue. Evaluate market potential using consumer spend trajectory alongside install volume to avoid over-investing in markets that deliver users but not buyers.
Most eCommerce teams use AI-powered analytics for quick-answer retrieval and single-metric lookups. Strategic queries involving anomaly detection, causal analysis, and multi-variable breakdowns represent a fraction of usage. Explore building workflows that prompt AI tools with layered, diagnostic questions rather than simple data pulls — especially during Q4 when the speed of identifying performance shifts determines outcomes.
About the authors
