The Background
StyleNest is a Delhi-based direct-to-consumer ethnic wear brand. They had been running Google Shopping and Meta Ads for about a year — spending ₹5 lakh a month — but their ROAS had flatlined at around 2.1x and their cost-per-purchase was climbing every month.
They came to us wanting two things: scale the budget without killing profitability, and reduce their dependence on discount-driven creatives — the constant "FLAT 40% OFF" ads that were training their audience to wait for sales rather than buy at full price.
The Real Problems We Found
Before touching a single campaign setting, we audited the entire account. Five structural problems were undermining everything:
- Google Shopping campaigns had no product segmentation — bestsellers and slow-movers were in the same campaign competing for the same bids
- Meta creatives were 90% discount-led — which tanked brand perception and trained the audience to expect permanent sales
- No proper remarketing funnel — cold traffic and warm audiences who had visited the site were being shown identical ads
- Attribution was last-click only — Meta's contribution to top-of-funnel brand awareness was completely invisible in the data
- Zero use of first-party data — no Customer Match lists, no lookalike audiences built from actual purchasers
What We Did
Google Ads — Restructure and Prioritise
We split campaigns into three clear tiers based on commercial priority: Hero Products (top 20 SKUs by revenue contribution), Seasonal (festive inventory and new arrivals), and Clearance (end-of-season stock that needed moving). Each tier got its own budget allocation and ROAS target — Hero at 5x, Clearance at 2x.
We added Performance Max campaigns layered with audience signals built from existing purchaser data — giving Google's algorithm the context it needed to find similar high-value customers. Price extensions and promotion extensions were restricted to the Clearance tier only. Full-price products were advertised at full price.
Meta Ads — Full-Funnel Creative Strategy
The single most important structural change was separating cold, warm, and hot audiences into distinct campaigns with different objectives and budgets — and building creative specifically for each stage.
Top-of-funnel (cold traffic) received lifestyle UGC content — real customers wearing StyleNest products at weddings and festivals. Middle-of-funnel received product education content — fabric quality explainers, sizing guides, how-to-style Reels. Bottom-of-funnel retargeting used social proof — customer reviews, trust signals, and a low-pressure call to action. Discount-led ads dropped from 90% of the creative mix to 15% — and were restricted to BOF retargeting only.
Data and Attribution Fix
We switched from last-click to Data-Driven Attribution in Google Ads, which immediately gave a more honest picture of which campaigns were actually driving purchase decisions versus which were just showing up at the end of a longer journey. We implemented Meta's Conversions API via Google Tag Manager — improving signal quality by 34% and restoring visibility into conversions that iOS had made invisible.
We created a 180-day purchaser list and uploaded it to both Google and Meta — using it to exclude existing customers from acquisition campaigns and build lookalike audiences that matched their profile.
The Numbers — Before vs After (90 Days)
| Metric | Before | After 90 Days | Change |
|---|---|---|---|
| Monthly Ad Spend | ₹5,00,000 | ₹20,00,000 | 4x Scale |
| Blended ROAS | 2.1x | 3.8x | +81% |
| Cost per Purchase | ₹1,240 | ₹820 | −34% |
| CTR (Google Shopping) | 0.9% | 2.3% | +156% |
| Meta CVR (Cold Traffic) | 0.8% | 1.6% | +100% |
| Revenue (Monthly) | ₹10.5L | ₹76L | +623% |
| New Customer % | 44% | 61% | +17 pts |
Key Takeaway: The single biggest lever was creative diversification. When you stop training your audience to buy only during discounts, your blended margin improves even as spend scales. The campaign restructure mattered — but the creative strategy is what moved the needle. The lesson carries into every e-commerce account we touch.