The Problem: ₹15L Monthly Ad Spend, 1.2x ROAS
When this health supplement brand came to us, they were spending ₹15 lakh per month on Google and Meta Ads and barely breaking even. Their previous agency had been running campaigns for 8 months with minimal improvement.
The brand's leadership was close to shutting down their paid media entirely.
Here's exactly what we found — and what we did about it.
The Audit: Where the Money Was Bleeding
Google Ads Issues
- 67% of budget going to broad match keywords with zero commercial intent
- No conversion value tracking — only counting leads, not revenue
- Shopping campaigns not using product feed optimisation
- No brand protection campaigns — competitors hijacking their brand name
- PMAX campaigns with generic asset groups cannibalising search campaigns
Meta Ads Issues
- All 14 ad sets targeting interests — no lookalike audiences
- Creative testing fatigue — same 3 creatives running for 5+ months
- No retargeting separation between warm and hot audiences
- No WhatsApp conversion tracking
- Broad targeting without demographic refinement
Landing Page Issues
- Homepage as landing page for all campaigns (catastrophic for conversion)
- Mobile page speed score: 34/100
- No social proof above the fold
- Add-to-cart to purchase conversion rate: 8% (industry average: 22%)
The 90-Day Restructure Plan
Week 1–2: Fix the Foundation
Before touching budgets, we:
- Implemented proper conversion value tracking — connected purchase value to Google Ads
- Fixed Core Web Vitals — mobile speed score went from 34 to 81 in 10 days
- Built dedicated landing pages for each top product — single-purpose, fast loading
- Added WhatsApp conversion tracking using Meta CAPI
Week 3–4: Google Ads Restructure
We paused all underperforming campaigns and rebuilt from scratch:
New Campaign Architecture:
- Brand Protection Campaign (exact match, low budget)
- Competitor Conquest Campaigns (separate budget)
- High-Intent Search — exact and phrase match only
- PMAX — with proper asset groups per product category and negative keyword lists
- YouTube for Upper Funnel — educating the audience
Key decision: We switched to Target ROAS bidding only after accumulating 30+ conversions per campaign per month. Before that, Maximize Conversion Value.
Week 5–6: Meta Ads Overhaul
Audience Architecture:
Cold (40% of budget):
- 1% lookalike of purchasers (last 180 days)
- 1% lookalike of high-value purchasers (top 25% by order value)
Warm (35% of budget):
- Video viewers 50%+ (last 60 days)
- Instagram engagers (last 30 days)
- Website visitors who didn't add to cart
Hot (25% of budget):
- Add-to-cart abandoners
- Checkout initiators
- Past purchasers (cross-sell/upsell)
Creative Strategy:
16 new creative variations produced in month 1:
- 4 UGC-style video testimonials (best performers)
- 4 product benefit reels (30 seconds)
- 4 educational carousels
- 4 offer/discount statics (surprisingly worst performers)
Week 7–10: Optimise and Scale
- Identified Sunday 7–10pm as highest conversion window — shifted budget accordingly
- Found women 28–42 in metros had 3x the LTV of other segments — increased bids
- Discovered WhatsApp conversions being incorrectly attributed — fixed attribution model
- Cut bottom 30% of ad sets by ROAS, reallocated to top performers
Results: Month 0 to Month 3
| Metric | Before | Month 1 | Month 2 | Month 3 |
|---|---|---|---|---|
| Monthly Spend | ₹15L | ₹15L | ₹16L | ₹18L |
| Revenue | ₹18L | ₹34L | ₹54L | ₹74L |
| ROAS | 1.2x | 2.3x | 3.4x | 4.1x |
| CAC | ₹1,240 | ₹720 | ₹540 | ₹410 |
| Conversion Rate | 1.1% | 2.3% | 3.1% | 3.8% |
Additional revenue in 90 days: ₹92 lakh Agency fees in same period: ₹5.4 lakh ROI on agency investment: 17x
5 Key Lessons
- Fix tracking before fixing campaigns — you're flying blind without proper data
- Landing pages matter more than ads — we saw bigger improvement from CRO than any single ad change
- Audience structure is everything — separating cold, warm, and hot audiences 3x'd Meta efficiency
- UGC always wins — our highest-performing ad cost ₹3,000 to produce (a WhatsApp video from a happy customer)
- Attribution is broken by default — most brands are making decisions on bad data
Is your ad spend stuck in similar patterns? Book a free performance marketing audit — we'll audit your accounts and show you exactly where your budget is leaking.