One of our D2C partners, a phone case brand running ₹30L/month in ad spend, was stuck running a single, undifferentiated Meta Ads campaign trying to do three separate jobs at once — awareness, consideration, and conversion — for every visitor regardless of where they actually stood in the buying decision. The fix was a 3-layer funnel restructure (TOF, MOF, BOF), paired with a timed WhatsApp and email automation layer for cart recovery. The result: a 4.5x ROI on the exact same monthly budget, with no increase in spend. This case study breaks down the full story — the problem, the discovery, the fix, and the metrics — structured so you can apply the same diagnosis to your own account.
monthly ad spend — unchanged before and after the fix
ROI on the exact same monthly budget
of abandoned carts recovered by the new automation sequence
Problem: One Campaign, Three Jobs
When we first audited this account, the setup looked simple on the surface: one primary campaign, reasonable targeting, decent creative, ₹30L/month in spend. But a single campaign was being asked to do three fundamentally different jobs simultaneously:
- 1Introduce the brand to people who'd never heard of it (a cold, top-of-funnel job)
- 2Build consideration with people who'd engaged but hadn't purchased (a middle-of-funnel job)
- 3Close the sale with people who'd already shown high intent — cart abandoners, repeat visitors (a bottom-of-funnel job)
One campaign, one audience setting, one message, trying to serve all three jobs at once. The result was predictable: mediocre performance across the board, because the messaging that might convert a high-intent cart abandoner was too aggressive for a cold visitor who'd never heard of the brand, and the softer brand-introduction messaging that might work on cold traffic did nothing to close warm, ready-to-buy visitors.
The Discovery
Running the account's data through a proper funnel-stage audit surfaced the core issue clearly: cost per acquisition was inconsistent and unpredictable month to month, not because of seasonal demand shifts, but because the single campaign's performance depended entirely on whichever audience segment happened to be more responsive that week. There was no structural reason the account should perform consistently, because it wasn't actually targeting anyone consistently — it was targeting everyone, identically, all the time.
A second, equally important discovery: once a visitor added a product to cart and left without purchasing, almost nothing happened next. A single generic retargeting ad, run at the same frequency regardless of how recently or how seriously that visitor had shown intent. No structured recovery sequence. No urgency signal timed to actual behavior. A meaningful share of near-purchases were simply evaporating with no attempt to recover them.
The Solution: The 3-Layer Fix (TOF / MOF / BOF)
The fix was structural, not creative. We rebuilt the single campaign into three distinct, properly separated layers:
Top-of-Funnel (TOF)
Broad and lookalike prospecting focused purely on brand introduction and product storytelling — no urgency, no hard sell.
Middle-of-Funnel (MOF)
Retargeting for engaged-but-not-purchased visitors, using testimonial and comparison content to build trust a cold audience hasn't developed yet.
Bottom-of-Funnel (BOF)
Retargeting for cart abandoners and high-intent visitors — social proof plus a direct call to complete the purchase, the only layer where urgency was appropriate.
Critically, each layer excluded the others — TOF excluded anyone who'd already engaged with the site, MOF excluded existing purchasers, and BOF targeted only the highest-intent segment specifically. Without these exclusions, the three layers would have quietly competed against each other in Meta's auction, inflating costs across the board. The full mechanics of this structure are covered in our complete full-funnel Meta Ads framework for D2C brands.
The Automation Layer
Restructuring the ad funnel alone wasn't enough — a properly segmented BOF campaign still needed somewhere real to send recovered intent. We built a structured, timed cart recovery sequence across email and WhatsApp: a gentle reminder at 1 hour, an urgency-based WhatsApp message at 24 hours, and a final price-anchored offer at 48 hours. This sequence alone recovered 35% of abandoned carts — full mechanics and message-by-message breakdown covered in our cart abandonment recovery playbook.
The ad restructure brought the right traffic to the right message. The automation layer made sure that traffic didn't quietly leak away again once it reached the site.
Results: Before and After
- 1Campaign structure: 1 undifferentiated campaign → 3 segmented layers (TOF/MOF/BOF)
- 2Monthly ad spend: ₹30L → ₹30L (unchanged)
- 3Overall ROI: Baseline → 4.5x
- 4Cart abandonment recovery: Ad hoc, single generic retarget ad → Structured 3-message sequence, 35% recovery rate
- 5Audience targeting consistency: Inconsistent, week-to-week variance → Stable, stage-specific targeting
- 6Budget allocation: Weighted toward broad reach → Rebalanced toward properly segmented BOF and MOF
The headline number is the 4.5x ROI, but the more important structural finding is the last row: none of this required more budget. It required the same ₹30L, reallocated toward the traffic that was actually ready to convert, instead of spread evenly and inefficiently across every stage of intent at once.
Growth Intelligence
One framework. Every two weeks.
No theory, no filler. Frameworks from live campaigns — directly in your inbox.
The Lesson for Every D2C Brand
The mistake in this account wasn't the product, the creative quality, or even the targeting settings within the single campaign — it was structural. One campaign cannot simultaneously introduce a brand to a stranger and close a sale with someone who already has three items in their cart. Every D2C brand running a single, undifferentiated campaign is very likely making the exact same structural mistake, regardless of how good the individual ad creative looks.
The second lesson sits downstream of the ad account entirely: even a perfectly built funnel loses value if nothing happens once a visitor reaches the site and doesn't convert immediately. The automation layer — not just the ad restructure — is what actually captured the value the funnel fix made possible.
This same underlying pattern — one undifferentiated process trying to serve fundamentally different customer stages — shows up constantly across the D2C accounts we audit. If your ad account looks similar to what this partner started with, our D2C customer retention guide covers what to build once the acquisition funnel itself is fixed.
Frequently Asked Questions
Meaningful improvement typically shows within the first 4-6 weeks of restructuring, once retargeting audiences for the MOF and BOF layers have had time to build to a workable size for stable performance.
No. The ₹30L/month budget remained unchanged throughout. The 4.5x ROI improvement came entirely from reallocating the existing spend across a properly segmented 3-layer funnel structure, not from spending more.
Both mattered, but they solved different problems. The funnel restructure fixed who saw which message. The automation layer fixed what happened after a visitor showed intent but didn't convert immediately — removing either one would have left significant value uncaptured.
Yes, the principle scales down — even a modest budget benefits from separating cold, warm, and hot audiences into distinct campaigns with appropriate messaging for each, rather than running one undifferentiated campaign regardless of total spend level.
The two metrics measure different parts of the funnel — cart recovery specifically measures BOF-stage conversion, while the 4.5x ROI reflects the full account's performance across all three funnel layers combined. Both improvements came from the same underlying restructure and reinforced each other.
Audit the current campaign structure first — specifically checking whether cold, warm, and hot audiences are currently being served the same message. This single diagnostic, covered in our CX audit framework, usually reveals whether a funnel restructure like this one applies.