When a fashion accessories brand from Surat approached me in early 2025, they were spending ₹2.5 lakh/month on Google Ads and getting a 2.1X ROAS. They knew the numbers weren’t good — but they’d been told by their previous agency that “2X is normal for accessories.” It isn’t. Within 90 days, we were at 8X ROAS on the same ₹2.5L budget. Here’s exactly how we did it.
The Situation When I Took Over
Before rebuilding anything, I ran a full audit. What I found was textbook under-performance:
- Campaign structure: One broad Shopping campaign with all 340 SKUs lumped together, zero segmentation by margin or price point
- Search campaigns: Broad match keywords with no negative keyword list — they were paying for searches like “jewellery repairing near me” and “free earrings”
- Bidding: Maximize Clicks — completely wrong objective for an e-commerce brand focused on ROAS
- Product feed: Titles were manufacturer-style (“SKU-4821-GLD-EAR”) with no searchable keywords, no GTINs, and incorrect product categories
- Conversion tracking: Only “Purchase” was tracked — no micro-conversions like Add to Cart or Initiate Checkout, which starved the algorithm of learning signals
The account had ₹7.5L in historical spend with poor data quality. Before touching the campaigns, I needed to fix the foundation.
Phase 1 (Weeks 1–2): Feed Overhaul and Tracking Fixes
The most impactful Google Shopping changes happen in the product feed — not in the campaign settings. We rewrote every product title following this structure:
[Material] + [Product Type] + [Key Feature] + [Gender/Occasion] + [Brand]
So “SKU-4821-GLD-EAR” became “Gold Plated Jhumka Earrings for Wedding — Traditional Indian Design — [Brand]”. This alone improved impression share by 34% within 10 days because the feed now matched actual search queries.
We also fixed conversion tracking — adding Add to Cart and Checkout Started as secondary conversions, with Purchase as the primary. This gave Google’s algorithm 6x more signals to work with during the learning phase.
Phase 2 (Weeks 3–6): Campaign Restructure
I split the single Shopping campaign into four segmented campaigns based on product margin and price tier:
- Campaign A — High Margin Hero Products (₹799–₹1,499): Top 40 SKUs by margin percentage. Budget: ₹80,000/month. Target ROAS: 10X
- Campaign B — Mid-Tier Volume Products (₹299–₹799): 120 SKUs driving most of the order volume. Budget: ₹1,00,000/month. Target ROAS: 7X
- Campaign C — Clearance / Low Margin: Remaining SKUs. Budget: ₹40,000/month. Target ROAS: 4X (acceptable at lower margin)
- Campaign D — Brand Search: Exact match brand terms only. Budget: ₹30,000/month. This campaign consistently delivered 15X+ ROAS
For Search campaigns, we replaced broad match chaos with a tightly controlled exact + phrase match structure, built a 400-term negative keyword list in week one, and added audience overlays (past purchasers, cart abandoners) to Shopping campaigns to improve bid efficiency.
Phase 3 (Weeks 7–12): ROAS Optimisation and Scaling
By week 6, Campaign A was already hitting 9.2X ROAS and Campaign B was at 6.8X — both ahead of targets. The restructure and feed work were paying off.
In phase 3, we focused on:
- Search term mining: Weekly review of search terms to find high-converting queries to add as exact match keywords
- Performance Max testing: We ran a PMax campaign alongside Shopping for the top 20 products to test incremental reach — it underperformed Shopping by 2.1X so we paused it at week 10
- Remarketing integration: Google Display remarketing targeting cart abandoners with product-specific creatives — this added ₹18,000 in incremental revenue/month at ₹6 cost per click
- Seasonal bid adjustments: Increased bids 20–30% on weekends and around payday dates (25th–31st of each month) — Indian consumer data consistently shows higher conversion intent during these windows
The Results at 90 Days
| Metric | Before (Month 1) | After (Month 3) | Change |
|---|---|---|---|
| Monthly Ad Spend | ₹2,50,000 | ₹2,50,000 | Same budget |
| Revenue from Google Ads | ₹5,25,000 | ₹20,00,000 | +281% |
| Blended ROAS | 2.1X | 8.0X | +281% |
| Average CPC | ₹18.40 | ₹11.20 | -39% |
| Conversion Rate | 1.2% | 3.8% | +217% |
| Orders/Month | 286 | 1,089 | +281% |
| Cart Abandonment Rate | 78% | 61% | -17pp |
The brand went from ₹5.25L in Google Ads revenue to ₹20L — same budget, same products, same website. The only things that changed were feed quality, campaign structure, bidding strategy, and conversion tracking.
What Made the Biggest Difference
If I had to rank the interventions by impact:
- Product feed rewrite (40% of total improvement): Google Shopping is a feed-first platform. If your titles and attributes don’t match real search queries, no amount of campaign optimisation compensates for it.
- Campaign segmentation by margin (30%): Treating a ₹299 anklet and a ₹1,499 statement necklace as the same product in the same campaign is a guaranteed way to average out your ROAS into mediocrity.
- Switching to Target ROAS bidding (20%): Once we had enough conversion data (90+ conversions in 30 days), switching from Maximize Clicks to Target ROAS unlocked Google’s algorithm properly.
- Negative keyword discipline (10%): A clean negative list prevents budget bleed to irrelevant searches. We added 400 negatives in week one and continue to add ~20/week.
Is 8X ROAS Achievable for Your Brand?
8X isn’t a universal target — it depends on your product margins, average order value, and competitive landscape. For fashion accessories with 60–70% gross margins and average orders around ₹650, 8X is achievable and sustainable. For a brand selling products at ₹200 AOV with 30% margins, the math is different.
But almost every Indian e-commerce brand spending on Google Ads has significant, recoverable ROAS leakage — usually in the feed, the campaign structure, or the tracking setup. The floor is almost always higher than where you currently are.
Running ads and not getting results? Book a free 30-minute strategy call — I’ll audit your account for free.
