What do you do when a Google Ads account looks healthy on the surface but refuses to grow? That was the situation with one of our health supplement clients. For 8 straight months, the account sat at roughly $30,000 per month in Google Ads revenue with a flat 3.0x ROAS. Spend was steady. Conversion tracking was clean. Creative was acceptable. Yet growth had stalled.
1) The situation
Why was the account stuck if nothing looked obviously broken? Because “not broken” and “ready to scale” are two very different things. This brand had solid products, repeat purchase behaviour, and enough historical conversion data to let Google’s AI work. But revenue had plateaued, and high-intent Shopping visibility was too low to unlock the next level.
We see this a lot with ecommerce brands, you know: campaigns get blamed first, while the product feed quietly limits performance in the background.
2) The diagnosis
So what did we find when we audited the feed? Three issues jumped out immediately:
- 34 products had missing GTINs
- 0 custom labels were in use
- Descriptions were generic and under 100 characters
That combination matters more than most brands realise. Google’s AI can only match products to nuanced search intent if the feed gives it enough context. If a listing just says “Ashwagandha Supplement” with a thin description, it has very little to work with compared with a richer product profile.
In this account, the AI wasn’t failing. It was under-informed.
3) The fix
What changed? We rebuilt the feed from the ground up.
First, we added GTINs to all 34 products so Google could identify inventory accurately and improve match quality across Shopping placements. Next, we created 5 custom labels to give the algorithm commercial context:
- Margin tier
- Seasonality
- Bestseller status
- Customer segment
- New vs. existing
Then we rewrote every title and description with intent-rich language. Instead of vague naming, we used formats like:
“organic ashwagandha root extract for stress relief — 60 vegan capsules”
That small shift gave Google clearer signals around ingredient, use case, format, and audience. Bear in mind, we did not launch a brand-new campaign structure here. The core change was feed quality.
4) The 60-day result
What happened over the next 60 days? The account moved fast:
| Metric | Before | After |
|---|---|---|
| Google Ads revenue | $30k/mo | $100k/mo |
| ROAS | 3.0x | 4.7x |
| Impression share on high-intent queries | 28% | 64% |
That means monthly revenue increased by $70,000, and across the 60-day window the uplift totalled roughly $140,000 in additional revenue run-rate impact. Just as importantly, efficiency improved at the same time. We weren’t buying growth with weaker economics. We were giving the system better inventory intelligence.
5) The key insight
So which change mattered most? The single biggest lever was the custom label for margin tier.
Once Google could distinguish products with 50%+ margins from those closer to 20% margins, the AI naturally started bidding more aggressively on inventory that could absorb higher acquisition costs. That changed the economics of the whole account. Instead of optimising purely around surface-level conversion signals, the system had a clearer view of what was actually worth scaling.
Here’s the simplest way to think about it:
Before: Google saw products as mostly equal.
After: Google knew which products were profitable enough to win harder on.
If your account is plateaued, that’s the takeaway we’d want you to act on first: audit the feed before touching bids, budgets, or campaign naming conventions. Start with GTIN coverage, then build custom labels that reflect how your business really makes money.
How much commercial context is your feed giving Google right now? And if we looked at your top 20 products today, would the AI know which ones deserve the strongest bids?
