by Austin LeClear
on July 1, 2026
TLDR: We restructured a lifestyle e-commerce brand’s Google Ads account by cutting Performance Max spend by 65% and moving that budget into Shopping and non-brand Search. Same ROAS profile. $1.3M more in revenue on just $64K more in spend. The whole play came down to letting Shopping do cold customer acquisition and using Performance Max to close warm traffic — a feeder strategy, not a Performance Max strategy.
That’s it. No new platforms. No new creative engine. No “AI-powered” anything. Just a serious rebuild of the campaign structure they were already paying for.
The interesting part is what we cut to make that happen. We pulled $233,000 out of Performance Max and put it somewhere it could scale.
Prefer video? Check out How I Made An Ecom Brand $1.3M More in 12 Months (With Only $60K More in Spend) on YouTube.
Still with us? Good — let’s dig in.
The Account We Walked Into
This is a lifestyle e-commerce brand. Lots of clothing, props, and a strong Halloween peak on top of a year-round subculture audience. The account looked healthy on paper: about $4.5M in total revenue out of Google Ads in 2024.
But healthy on paper isn’t the same as dialed in.
When we opened up the account, it was running five separate Performance Max campaigns, a handful of segmented shopping campaigns, and even a stray display test someone had forgotten to clean up.
In 2024 they spent $457,000 in total. Here’s where the money went:
- PMax Campaign: $358,000
- Shopping: $50,000
- Search: $45,000 (roughly $17,000 of that was brand)
Roughly 78% of the entire budget was funneling into Performance Max. Their non-brand search campaign got about $28,000 for the whole year — the budget of a serious side project, not a serious growth lever.
The bigger issue was structural. Performance Max needs around $50 a day in spend per campaign to function. Below that, smart bidding doesn’t get enough conversion volume to learn, and the campaign slowly suffocates. Five separate Performance Max campaigns means you’re forced to pump at least $250 a day just to keep five algorithms breathing. The structure itself was making it impossible to put real money behind anything else.
The account worked, but it was leaving a lot on the table.
Why Performance Max Was the Bottleneck
Performance Max can quietly turn into a warm-traffic recycler.
Once it finds a pocket of people who already know your brand — people who’ve been on the site, hit the cart, and bounced — Google keeps going back to that pocket. That’s where the easy conversions live. That’s how a PMax campaign keeps its ROAS looking pretty while real customer acquisition stalls.
There’s a second issue once you try to scale past a certain ceiling. When you push more ad spend into a Performance Max campaign that’s already plateaued, Google doesn’t go find fresh cold shoppers for you. It spends the extra money on display remarketing and cold display inventory because those are the cheapest impressions in the system. Your new spend re-touches people you already had. The dashboard shows more conversions. The business sees the same customers.
Then there’s the overlap problem. With five Performance Max campaigns running at once, the same products were showing up in multiple campaigns and starving each other of conversion data. Every campaign bidding on the same inventory makes the algorithm worse at learning, not better, because the conversion signal gets split five ways. Each product should live in exactly one Performance Max campaign.
Read more: We Cut Performance Max Spend by 80% (And Still Hit Our Goals)
Here’s the part most accounts miss. For an e-commerce client, we only really care about one number under the hood: new customer acquisition. Repeat buyers and branded search terms were going to convert anyway. If most of Performance Max’s “wins” are coming from people who were already going to buy, campaign performance looks fine but you’re not growing the business — you’re paying Google a tax on customers you already had.
That’s exactly what was happening here. Five Performance Max campaigns chewing through $358K in ad spend, but the new-customer needle wasn’t moving the way it should have for that spend level. The math just stops working.
How We Restructured It
We rebuilt the account around a simple principle: about 80% of e-commerce revenue on Google Ads flows through shopping ads anyway. So let shopping campaigns do the heavy lifting on new customer acquisition, and let Performance Max scoop up the warm traffic that shopping already heated up.
That’s the feeder strategy in one sentence: shopping is the feeder, and Performance Max and search remarket and close.
We collapsed the whole Google Ads account down to this:
- Shopping – Top Sellers: Their top-performing brands, lumped into one shopping campaign. Not individual SKUs — brand-level portfolios pulled straight from the product feed.
- Shopping – High Margin Brand: One specific brand had unusually high margins compared to the rest of the catalog. It didn’t make sense to bury it inside the same campaign and bid against itself. Its own shopping campaign, its own bidding strategy, its own goals.
- One PMax campaign: Just one. Asset groups did the segmentation work that five separate Performance Max campaigns used to do badly.
- Search – Non-Brand All: A consolidated non-brand search campaign with tightly themed ad groups built around the search terms that were actually converting.
- Search – Non-Brand Deals: A separate search campaign we use for inventory blitzes, sales, and clearance pushes when they need to move product.
- Brand campaign: Untouched. Brand keeps printing money on its own.
That’s the whole structure. We went from a tangle of campaigns to a handful with clear jobs.
There’s a real mechanism inside this structure that makes it work, and it’s worth slowing down on.
The bidding logic: When you have the same product in both a Standard Shopping campaign and a Performance Max campaign, Google decides which one shows based on ad rank. Ad rank is driven by your bid. So if you set a lower target ROAS on the Standard Shopping campaign than on Performance Max, that sounds backward — until you remember a lower ROAS target means Google is allowed to bid more aggressively to hit it. Shopping wins the auction on the cold first click. Performance Max sits back and waits for the warmer return visit.
That’s the engine. Most bidding strategies get chosen by campaign type in a vacuum, but the real move is picking bid targets that make your shopping campaigns and PMax work as a system, not as competitors.
Campaign priority on the shopping side: Standard Shopping campaigns have a built-in High/Medium/Low priority setting that controls which campaign enters the auction first when a query is eligible across multiple shopping campaigns. We use it like this: High priority on the catch-all (a generic shopping campaign that covers everything else, paired with negative keywords); Medium on category or top-seller campaigns; Low on the high-margin brand and brand-specific campaigns. The query sculpting funnels traffic to the right place even though everything lives in the same product feed.
One Performance Max campaign, no product overlap: With one Performance Max campaign instead of five, asset group segmentation does the work that separate campaigns were trying to do badly. Each asset group focuses on a specific product line or audience signal, and no product is duplicated across campaigns. Every product lives in exactly one place. That’s how Performance Max actually learns.
Brand stays out of non-brand: We exclude branded search terms from Performance Max and from the non-brand search campaigns. Brand has its own dedicated campaign. If you let brand search bleed into your acquisition campaigns, brand returns will pull up the aggregate ROAS and make non-brand campaign performance look healthier than it is. Without clean conversion tracking on each campaign type, you’ll scale into a dead zone and wonder why it stopped working.
Read more: Google Shopping Optimization — Query Sculpting & Feed Management
Where the Money Was Sitting
Before we rebuilt the structure, we pulled every converting search term from 2024’s shopping campaigns and audited what was driving sales.
There was a clear gap.
The Google Ads account had a long list of non-brand terms that converted on shopping but had no dedicated search coverage. People searching for product categories and competing brand names — exactly the people you want to capture with a text ad and a tightly themed ad group — were not seeing search ads. Or worse, they were seeing them with $28K to fight over for an entire year.
We built S-Non-Brand-All off that data: one campaign, multiple themed ad groups grouped under similar bid goals. Consolidating let us pool conversion data so the bidding strategy could learn properly.
Two practical notes on how this gets built, because the audit step is where most accounts give up:
Where to mine the data: Two places. First, pull search terms from the existing Standard Shopping campaigns directly. Second, inside Performance Max, go to Insights and Reports → Search Term Insights → View Detailed Report. Performance Max hides individual queries by default, but the categorized themes are visible there. That’s where you find the non-brand terms Performance Max is converting on that you don’t have a search campaign for.
Why consolidate instead of split: Smart bidding needs roughly 30+ conversions per campaign per month to learn well, and 50+ if you’re running Target ROAS. If you split a search campaign into five tightly themed micro-campaigns, you’re starving each one of conversion data and the bidding strategy can’t optimize. Single keyword ad groups (SKAGs) are dead for the same reason. Themed ad groups inside one campaign, with the conversion data pooling at the campaign level, is the structure that learns.
We also started the new search campaign with exact and phrase match, then layered broad in later once smart bidding had enough conversion data to filter out junk. Broad match without proper conversion tracking is how google ads accounts light money on fire — Google will fire on anything semantically related, including search terms nowhere near purchase intent.
This is one of those moves that sounds boring until you see the numbers. Most accounts have this gap. They just don’t audit for it because they assume shopping is covering it.
The Results: $1.3M More on $64K Extra Spend
In 2025 we spent $521,000 total in ad spend, which is $64,000 more than 2024.
Here’s how the spend got redistributed across the account:
- Shopping campaigns: went from $50K to $300K, a roughly 500% increase, or $257K more in spend.
- Search: went from $45K to $100K, a roughly 100%+ increase, or $52K more.
- Performance Max: the PMax campaign spend went from $358K to $125K, a 65% cut, or $233K less.
The revenue picture moved the same way:
- Search revenue: up roughly $738,000 year-over-year.
- Shopping revenue: up roughly $2.1 million.
- Performance Max revenue: down about $1.6 million.
Same ROAS profile, campaign performance strengthened, and $1.3M more in the business.
Aggregate ROAS held steady at approximately 11x in 2025, in line with 2024 and well above this client’s target.
Same return profile. $1.3 million more in topline. $64K more in spend. That’s the whole story.
Why the Feeder Strategy Compounds
The reason this kind of rebuild compounds is the customer journey, not the campaign types.
Most people who click a shopping ad don’t buy on the first visit. They click, they browse for five minutes, maybe they add to cart, maybe they hit the checkout and stall out. Google sees all of that. Google now knows those people show high intent to convert.
When those same people come back the next day and search for the brand, or search for the product category, or open a YouTube tab on their phone, that’s when Performance Max and remarketing earn their keep. They scoop up the warm audience that shopping campaigns just heated up.
Without aggressive shopping pushing new traffic into the system, the PMax campaign has nothing fresh to chase. It just keeps cycling through your existing brand audience and calls it growth.
With aggressive shopping doing the customer acquisition work, Performance Max becomes a closer instead of a recycler. Same campaign type, completely different role.
Read more: Google Ads Feeder Strategy — +109% Sales Increase in 22 Days
That part is hard to see if you’re only looking at Performance Max in isolation. The campaign isn’t broken — it’s just being asked to do a job it isn’t built to do at scale.
Should You Run This Play in Your Account?
Probably, but not blindly.
This rebuild works best when:
- You’re an e-commerce advertiser doing meaningful shopping volume already.
- Your Performance Max campaigns are taking 60%+ of total ad spend and you can’t explain why scale stops working past a certain point.
- You can see brand and repeat-customer traffic baked into your Performance Max campaign performance.
- You have non-brand search volume that’s converting on shopping but doesn’t have dedicated text-ad coverage.
- You can sustain at least 30 conversions per month per active campaign so smart bidding has enough signal to learn.
- You care about customer acquisition, not just ROAS.
This is not the right move when:
- You’re a small e-commerce brand under a few thousand a month in ad spend, with no real shopping campaigns volume to feed anything.
- Your product feed is in rough shape. A feeder strategy with a broken feed just spends faster.
- You’re in a category where 90% of the conversion path lives in YouTube or Discovery, not Search.
- You don’t have a way to verify customer acquisition outside the Google Ads account dashboard.
Read more: Stop Optimizing for ROAS — Break Through Your Growth Ceiling
One non-negotiable for the last point. If you’re moving budget out of Performance Max, set up a third-party attribution tool first — Wicked Reports, Triple Whale, or NorthBeam, depending on stack. Google’s in-platform attribution will tell you what Google wants you to believe about its own performance. You need an outside source of truth that can show you the new customer percentage on every campaign type. Otherwise you’re trusting the campaign type you’re trying to demote to grade its own homework.
If most of that list matches your account, this is the rebuild to run.
Final Thoughts
This is not complicated work. There’s no secret bid strategy hiding in the screenshots.
We took an account that was over-segmented on Performance Max, found the search gap, consolidated the structure, and pushed budget toward the campaign type that scales for e-commerce. That’s it.
The bottom line: $1.3 million more in attributed revenue, $64,000 more in spend, same return profile, happier client.
Most accounts we audit have some version of this same problem: too much weight on Performance Max, not enough discipline on shopping structure, and a search gap nobody bothered to map. Fix those three and the math usually starts working in your favor in 60 to 90 days.
Frequently Asked Questions
Why is Performance Max not working in my account?
Performance Max often stops working when it takes 60%+ of your ad spend. Past that ceiling, Google spends the extra budget on display remarketing instead of finding new customers, so your dashboard shows more conversions but the business sees the same customers. The fix is structural — cut Performance Max spend, push budget into Shopping campaigns for cold acquisition, and let Performance Max close warm traffic.
How much should I cut Performance Max by?
In this case study we cut Performance Max by 65% — from $358K to $125K in annual spend. The right cut depends on your account. A good rule of thumb: if Performance Max is taking more than 60% of your total ad spend and new customer acquisition has flatlined, you’re probably over-invested in PMax and should redirect budget to Shopping campaigns.
What is a feeder strategy in Google Ads?
A feeder strategy means Shopping campaigns do the heavy lifting on cold customer acquisition, while Performance Max closes the warm traffic that Shopping already heated up. Instead of Shopping and PMax competing for the same auctions, they work as a system — Shopping wins the cold click, PMax converts the return visit.
How do you know if Performance Max is only re-touching existing customers?
Set up third-party attribution (Wicked Reports, Triple Whale, or NorthBeam) so you have an outside source of truth for new customer percentage on every campaign type. Google’s in-platform attribution can’t reliably tell you whether Performance Max is acquiring customers or just re-marketing to people your other channels warmed up.
Should I run one Performance Max campaign or multiple?
One. With multiple PMax campaigns, the same products get bid on across campaigns and the algorithm’s conversion signal gets split, which makes learning worse, not better. Use asset groups inside one PMax campaign to do the segmentation work that separate campaigns were trying to do badly.
Get a personalized, 1-on-1 Google Ads account review.
Want us to look at your account and tell you exactly where the gaps are?
We offer a free 30-minute account analysis for advertisers who qualify. It’s a custom-tailored, in-depth review of your Google Ads account that reveals day-one improvements to your campaigns — the kind of strategy we used to add $1.3M in revenue to the account in this case study.
We don’t take every account that applies. We work with businesses where we’re confident we can deliver real results, so we’ll need a few details on your business and goals before we book the call.
If that sounds like a fit, click below to apply.
