Why Is My Mobile ROAS 4x but Desktop ROAS Negative: Device-Split Diagnosis for D2C

Your mobile ROAS is 4.1x. Your desktop ROAS is 0.6x. Same campaign, same creative, same offer. Meta's CBO doesn't care — it lumps the spend together and reports a blended 2.8x.

For Indian D2C, this device split is almost universal. Mobile drives 92-96% of traffic in India and almost all impulse purchases. Desktop catches a different shopper — usually higher AOV, longer consideration — and most ad accounts aren't built to serve that buyer well.

First: Confirm the Split Is Real, Not Just a Sample Size Issue

Desktop is often only 4-8% of impressions. A few bad days can show as a 'crisis' when it's noise.

• Require 30+ days and 1,000+ desktop impressions before treating the split as signal.

• Check spend share. If desktop is 5% of spend, even a 0.5x ROAS is only a tiny absolute leak.

• Confirm currency display. Some accounts double-count cross-device — see our [INR vs USD guide](https://www.wittelsbach.ai/post/inr-vs-usd-currency-confusion-in-meta-ads-dashboards-and-the-fix).

The Root Cause: Different Buyer, Different Funnel

Mobile users in India are scroll-buyers. They tap, read 30% of the PDP, add-to-cart, and check out in under 4 minutes. Desktop users research. They compare. They open 3-5 tabs. They leave and come back two days later.

If your ad and PDP are mobile-first — short copy, large hero image, single CTA — desktop visitors hit a page that looks empty and quit. Meta still optimises against blended data and keeps pushing budget toward desktop placements that don't convert.

• Creative is sized for vertical 9:16 so desktop sees letterboxed videos.

• PDP has no comparison or detailed content so desktop researchers bounce.

• Checkout is mobile-optimised so desktop checkout feels cramped or buggy.

• Retargeting windows are short so desktop's longer consideration cycle isn't captured.

The 4-Step Device-Split Diagnostic

Step 1: Pull Device Breakdown by Spend and Revenue

Ads Manager → Breakdown → By Delivery → Impression Device. Capture spend, purchases, AOV, and ROAS for each device. Most Indian D2C accounts find desktop is 5-10% of spend but contributes less than 2% of revenue.

Step 2: Compare Funnel Drop-off

Pull LPV → ATC → Purchase rates per device. If desktop's LPV→ATC rate is half of mobile, the issue is the page. If LPV→ATC is fine but ATC→Purchase craters, the issue is checkout.

Step 3: Audit Creative Display on Desktop

Preview your top ad in desktop Feed. If the text is cut, the CTA button isn't visible, or the video has heavy letterboxing — you've found the leak.

Step 4: Check AOV by Device

Desktop AOV is usually 1.4-2.0x higher than mobile in India. If your desktop AOV is equal or lower, you're failing to capture the high-intent researcher who normally spends more.

The Fix: Two Paths That Both Work

Path A: Exclude Desktop

If desktop is under 10% of spend and the leak is consistent, just exclude desktop placements at the ad set level. Recovers 4-9% blended ROAS instantly for most accounts. The downside: you lose the 1-2% of high-AOV desktop buyers.

Path B: Build a Desktop-First Funnel

If desktop is 15%+ of spend and you sell ₹3,000+ AOV products, build a dedicated funnel: a separate ad set with desktop-optimised creative (16:9 video, longer copy), a comparison-rich PDP, and 30-day retargeting windows. Higher effort, but desktop buyers in India often beat 5x ROAS when treated right.

How Wittelsbach AI Spots Device Leaks Without Manual Pulls

Bach AI automatically segments every campaign by device and surfaces accounts where one device is silently bleeding the blended ROAS. It quantifies the exact ₹ leak per month and recommends Path A or Path B based on your spend share and AOV. Try Bach AI on your account at [app.wittelsbach.ai](https://app.wittelsbach.ai).

Frequently Asked Questions

Should I always exclude desktop for Indian D2C?

No. For sub-₹2,000 AOV impulse products — apparel, snacks, mobile accessories — desktop excludes are usually correct. For ₹3,000+ AOV products, especially jewelry, home, electronics, beauty, and considered purchases, desktop buyers spend more and convert at higher AOV when served well. The right call depends on your category, not a blanket rule.

How much of total Indian D2C traffic is desktop in 2026?

Roughly 4-8% across most categories, but it skews higher for B2B-leaning D2C, premium home goods, and finance/insurance products. The number is shrinking year over year but the desktop buyer's AOV typically remains 1.5-2x mobile. So even at 5% of traffic, desktop can contribute 8-12% of revenue if served properly.

Can I run different creatives by device in a single ad set?

Yes — at the ad level you can specify placement customizations, including device-specific creative. For most Indian D2C brands the simpler move is a separate ad set per device, so you can also vary copy length, CTA, and landing page. Mixing device customizations inside a single ad set works but makes performance attribution and creative iteration messier.

Will excluding desktop affect Meta's algorithm or learning?

Marginally. Desktop is a small share of impressions, so excluding it doesn't trigger a learning reset. Expect a 24-48 hour redistribution as Meta shifts your budget to mobile placements. After that, you're back to steady state — usually with slightly higher blended ROAS because the dead weight is removed.

Is checkout breakage the most common desktop issue?

For Shopify and WooCommerce stores in India, the most common desktop issue is actually creative letterboxing followed by PDP layout breaking on wider screens. Checkout is usually fine. The fixes that move the needle: a 16:9 hero creative variant, a desktop-specific PDP grid that uses the wider viewport, and a longer retargeting window for desktop researchers.