← All articles

Mobile Ecommerce Statistics: Traffic Share vs Revenue Share

November 16, 2026

Mobile Ecommerce Statistics: Traffic Share vs Revenue Share

Smartphones generated 56.4% of U.S. online holiday spending in Adobe's 2025 analysis, while its 2026 forecast puts that share at 57.4%. Neither figure is mobile traffic share. They describe the share of spending attributed to a device, which depends on how often visitors buy and how much they spend when they do.[1][2]

A mobile-first shopping journey can still produce a different device mix at checkout, during a return or on a later visit. To make mobile ecommerce statistics useful, match the metric to the business question. Traffic helps explain where attention arrives. Revenue shows where measured purchasing value lands.

Mobile dominates holiday spending, with a different share for different activities

The following findings come from Adobe's U.S. retail analysis. They are not global annual averages and should not be relabeled as the share of all ecommerce conducted on phones. The completed season covers November and December 2025; the following season remains a forecast in the cited September 2026 publication.

Selected Adobe holiday mobile-commerce measures
MeasureSharePeriod and statusWhat the denominator represents
Smartphone share of online spending56.4%2025 holiday season, reportedOverall online spending in the analysis
Mobile share of BNPL purchases82.2%2025 holiday season, reportedPurchases using the payment method
Mobile share of returns38.8%2025 holiday season, reportedReturn activity in the analysis
Mobile share of online spending57.4%2026 holiday season, forecastProjected overall online spending

Sources: Adobe's 2025 season recap and 2026 forecast. The figures use different activity denominators; their difference is not a device-conversion calculation.

There is a useful operational signal here: the device used to buy is not necessarily the device used to manage the purchase afterward. A store should test product discovery, payment and post-purchase service separately rather than treating mobile optimization as a single checkout-screen project.

Traffic, order and revenue share answer different questions

Mobile traffic share is mobile sessions divided by all sessions within a defined measurement system. Order share is orders attributed to mobile divided by all attributed orders. Revenue share is the revenue attributed to mobile divided by total attributed revenue. These definitions are analytical formulas, not additional market observations.

Revenue share can differ from order share because order values differ. Order share can differ from session share because session conversion differs. Cross-device journeys can complicate both comparisons: a shopper may browse on a phone and later buy on another device, depending on how the reporting system assigns the order.

Choose a consistent revenue boundary. Gross sales, discounted sales and net sales after refunds will not always produce identical shares. If mobile orders have a different refund pattern, a gross-sales device chart may look different from a net-sales chart even when the original checkout events are unchanged.

Also define mobile consistently. Some tools separate smartphones and tablets; other dashboards group them. A smartphone-specific external statistic should not be compared directly with an internal mobile category that includes tablets without noting the difference.

A traffic-heavy device can generate a smaller revenue share

Consider an illustrative store with 10,000 sessions: 7,000 from phones and 3,000 from desktop. Assume phone visitors place 140 orders at an average value of $60, while desktop visitors place 120 orders at an average value of $90. These are hypothetical inputs, not a benchmark dataset.

Hypothetical device analysis with consistent session and order definitions
MetricPhoneDesktop
Sessions7,0003,000
Session share70%30%
Orders140120
Session conversion2%4%
Average order value$60$90
Revenue$8,400$10,800
Revenue share43.75%56.25%

The phone accounts for most visits but less than half of revenue. Both conversion and average order value contribute to the difference. Looking only at traffic would overstate mobile's share of measured sales; looking only at revenue would conceal how much shopping activity happens on the phone.

This does not prove that a desktop-oriented design is better. The traffic may have different intent, campaign origin or customer mix. The example shows which questions to investigate, not which device should receive the next advertising dollar.

Revenue per session makes the comparison more actionable

Revenue per session is device revenue divided by device sessions. In the illustrative example, phone revenue per session is $1.20 and desktop revenue per session is $3.60. These values combine conversion frequency and order value in a single measure.

A useful diagnostic index divides a device's revenue share by its session share. An index of one means the device generates the same revenue per session as the site overall. Below one means lower revenue per session; above one means higher. This is an arithmetic identity when the session and revenue sets use compatible boundaries.

For the phone in the example, 43.75% divided by 70% equals 0.625. The same result appears by dividing phone revenue per session, $1.20, by site revenue per session, $1.92. A marketer can therefore calculate the index from a share chart without mistaking it for a conversion-rate index.

Keep the name explicit. Revenue per session is not profit per session, and a high-value order can have a different product margin or return risk. If the decision is about budget allocation, connect the order values to contribution costs before judging profitability.

A device worksheet should separate diagnosis from explanation

Start with a period that is long enough for the store's own purchasing cycle and use identical dates across reports. Export sessions, orders, sales, discounts and refunds with the available device dimension. Record where device attribution comes from and whether unknown-device activity is excluded.

Then segment the obvious confounders. Compare new and returning customers, campaign types, country and major product groups where the data permits. A broad phone-versus-desktop gap may narrow when the groups contain similar intent. It may also persist and point toward a practical checkout or product-page issue.

Test the actual tasks rather than only checking a responsive layout. Can someone select a size, view shipping information, apply a discount, complete payment and find a return instruction on a small screen? Keep the observed failure attached to the test rather than inferring it from a conversion gap alone.

TrueMargin's recommended operating view is a device comparison linked to order economics. Use it to prioritize investigation. The external benchmark supplies context, while your own store records identify the opportunity that can actually be acted on.

Post-purchase device activity deserves a separate report

A phone checkout and a desktop return are not necessarily different customers. They can be different stages of one customer's journey. If your reporting assigns each event to the device that generated it, the return device should not overwrite the original order device.

Keep an order-level view that preserves checkout device and adds the later service event separately. That allows a merchant to distinguish where the purchase happened from where help was requested. It also prevents a return portal's device mix from being misread as a product-quality comparison between mobile and desktop shoppers.

For a practical review, test the path from an order confirmation email to tracking and returns on both devices. Record the specific task that fails. The operational finding is more actionable than an unexplained percentage gap.

Keep the device categories stable during the comparison. If a tracking change moves tablet sessions between groups or increases the number of unknown devices, annotate that change before interpreting a trend. A measurement revision can alter a share chart even when the underlying shopping behavior has not materially changed.

Methodology and source boundaries

This article reconciles Adobe's published holiday findings and forecast with standard device-reporting formulas. Adobe describes its analysis as covering more than one trillion visits to U.S. retail sites, 100 million SKUs and 18 product categories.[1] That is the publisher's description of its coverage, not a claim that TrueMargin independently obtained the underlying records.

The January release uses both transaction and spending language around its headline mobile figure. The September forecast explicitly compares the earlier figure with overall spending, so this article labels that comparison as spending share. It does not convert the value into an order-count share.

All worksheet values and derived indexes above are explicitly hypothetical. We have not measured a universal mobile conversion gap or claimed the example represents a typical merchant. The forecast should retain its forward-looking label until replaced by an appropriate completed-season source.

Frequently asked questions

Is mobile traffic share the same as mobile sales share?

No. Sales share depends on conversion and order value as well as the share of visits. Use a traffic statistic only to describe traffic.

Can I calculate mobile conversion from revenue share?

Not without compatible order and session counts. Revenue share divided by traffic share gives a relative revenue-per-session index.

Should tablets be included in mobile?

Use a definition consistent with the benchmark and your reporting tool. Label smartphones separately when the source does.

Does the forecast establish the final 2026 result?

No. A forecast is a planning estimate. Preserve its publication date and replace it with a final report when one is available.

Related research: see how refunds affect advertising revenue reports for a worked application of these measurement distinctions.

Sources

  1. Adobe: 2025 holiday shopping season recap, released January 2026.
  2. Adobe: 2026 holiday shopping forecast, released September 2026.