We've managed $120M+ in Google Shopping spend across 20+ brands.
Here's what we learned: the brands ranking highest on prospecting terms like "women's dress" or "creatine" aren't always the ones with the biggest budgets.
They're the ones who understand how Google's algorithm scores products. (It's not complicated. But it's rarely prioritized.)
We call this the F.I.L.L. Protocol: Feed, Images, Landing pages, Listings. Four pillars that determine whether you rank or get buried. (We learned this the hard way - years of expensive feed audits before we codified it.)
First, you need to understand what Google's algorithm cares about.
Think of it like a scoreboard. Every product ad gets a score. The ones with the best score, combined with a competitive bid, end up at the top.
That score comes down to:
Translation: Google won't put you in #1 even if you bid the most.
They only want to push ads that drive engagement and make the user happy.
Once you understand that, you can outrank brands throwing way more money than you. (We've watched $500k/mo brands get beaten by $50k/mo brands who optimized their feed.)
This is where you tell Google exactly what your product is so they know who to show it to.
Your feed is the foundation. Get this wrong and nothing else matters. (Seriously - we've seen accounts 3x performance from feed fixes alone.)
Your product title is the most important element in your feed.
What to include:
Structure by category:
| Category | Title Structure |
|---|---|
| Apparel | Brand + Product Name + Type + Color + Size |
| Electronics | Brand + Type + Size + Color + Carrier |
| Home & Garden | Brand + Type + Feature + Feature |
| Beauty | Brand + Product Name + Type + Size + Key Benefit |
| Supplements | Brand + Product + Form + Quantity + Key Attribute |
Example:
Google only displays the first 70 characters in most placements. Front-load your most important details.
High-quality images that stand out get more clicks. More clicks = higher expected CTR = better ranking.
Basic requirements:
Advanced optimization: Use AI tools to generate multiple image variations. Test different angles, backgrounds, and styles. We've seen CTR jumps of 40%+ from image testing alone. The algorithm will favor the variation that gets the most engagement.
Fill in every relevant attribute:
Required attributes:
Strongly recommended:
More attributes = better matching = more relevant impressions. We've seen accounts jump 15-25% in impression share just by filling in missing fields.
This optional attribute is often ignored. Don't skip it. (We find this empty in 70%+ of accounts we audit.)
Product types help Google understand your product hierarchy and improve search relevance.
Format: Start broad, get specific.
Example: "Apparel & Accessories > Shoes > Athletic Shoes > Running Shoes > Men's Running Shoes"
Provide GTINs whenever available.
Retailers who add correct GTINs see 20% more clicks on average. (Free improvement. Takes 5 minutes.)
Why? Google understands exactly what you're selling. They can add rich product information. Your ads appear in more places.
Go to Merchant Center. Click "Store Quality."
Google grades you on:
Fix the areas where you're lagging. Google tells you exactly what to do. (Most people never check this. Don't be most people.)
What Google wants:
What to do:
What Google wants:
What to do:
What Google wants:
What to do:
What Google wants:
What to do:
Product reviews boost CTR. Set them up properly.
Options:
Different from product reviews. These rate your store overall.
Enable through Google Customer Reviews or approved third-party providers.
If you run evergreen promos or any kind of sale, set up "Promotions" in Merchant Center.
Promo badges on Shopping ads increase CTR.
Google doesn't want users clicking an ad and bouncing in 2 seconds.
Your landing page needs to deliver on the ad's promise.
What to check:
Mismatches = bounces = lower landing page score.
Target under 2.5 seconds.
Use Google PageSpeed Insights to identify issues.
Common fixes:
They should immediately feel they're in the right place.
Include:
Over 60% of Shopping clicks are mobile.
Check:
Test different formats to improve conversion:
Keyword-specific landing pages: Built around exact search terms. "Best creatine for women" → dedicated page for women's creatine.
Sales pages: Long-form, focused on conversion. Works well for higher-consideration products.
Product images are critical for CTR. AI makes it possible to test dozens of variations. (This is where it gets fun.)
Go to Auction Insights in your top-performing campaigns and ad groups.
Look at:
Now you know who's competing for your most valuable keywords.
Search their keywords with a VPN or use Google's Transparency Center.
See what creatives, styles, and offers your competitors are running.
Feed those examples into Gemini or Claude for visual analysis.
Ask it to break down:
Start with one clean, front-facing, high-resolution image of your product.
This becomes your baseline for every variation.
Upload to Gemini with specific settings:
Example prompt: "Create a lifestyle photo of this water bottle in a gym locker room. Soft shadows. Stainless steel reflections. Gritty but clean."
For lifestyle images, describe exactly what you want. The more specific, the better.
Usually takes a few rounds to get it right. Use follow-up prompts to refine.
Deploy AI-generated variations across your ads.
Focus on:
What to watch: During tests, one variation usually gets most of the traffic. That's the algorithm telling you it prefers one image over others.
When you see that, double down. You've found your winner. (Google just did your creative testing for you.)
| Factor | Impact | How to Optimize |
|---|---|---|
| Product Title | High | Include search terms, follow category structure |
| Image Quality | High | Stand out from competitors, test variations |
| GTIN | Medium-High | Add wherever available (+20% clicks) |
| Attributes | Medium | Fill all relevant fields |
| Store Quality | Medium | Fix Merchant Center grades |
| Landing Page Speed | Medium | Target <2.5 seconds |
| Product Reviews | Medium | Enable and grow review count |
| Price Competitiveness | Medium | Monitor competitor pricing |
Get these right and you'll outrank bigger brands spending more money.
The algorithm rewards quality, not just budget.
The F.I.L.L. Protocol came from years of mistakes:
Mistake: Keyword-stuffing titles. Early on, we crammed every possible search term into titles. Google caught on. Ads got disapproved. Quality score tanked. Now we focus on natural, descriptive titles that include relevant terms without manipulation.
Mistake: Ignoring product types. We skipped this "optional" attribute for years. Massive mistake. Product types help Google understand your catalog hierarchy. Accounts that fill this properly see better matching and more relevant impressions. It's not optional - it's essential.
Mistake: Generic images. We used manufacturer images - the default playbook. Worked fine until competitors started testing AI-generated lifestyle variations. Their CTR jumped 40%+. Now we treat image testing as core optimization, not nice-to-have.
Mistake: Focusing only on feed. We optimized titles and attributes religiously but ignored Merchant Center quality scores. Then Google released the Store Quality dashboard and we realized landing page speed and return policies directly impacted ad rank. The F.I.L.L. Protocol now treats all four pillars equally.
Follow this protocol and you skip the expensive lessons we learned.
This framework is based on $120M+ in Google Shopping spend across 20+ brands. It's what the algorithm needs to match your products to the right queries and drive higher ROAS.
We built a checklist that covers every attribute, every Merchant Center setting, every image requirement. Run it on your feed in 30 minutes.
Reply "FEED" and I'll send it over.
(Must be following)