Is Performance Max good or bad for ecommerce brands?

It's a powerful tool that defaults to flattering itself. Left unstructured, PMax leans on branded search and remarketing to inflate its reported return. Structured properly — brand exclusions, margin-based segmentation, clean feed data, real audience signals — it's one of the best prospecting engines Google has ever shipped. The difference is in the setup, then in the filter phase after launch: combing search terms for the intent the algorithm misread and cutting it with negative keywords as it surfaces.

Why is my Google Ads ROAS high but my profit flat?

Almost always attribution, not performance: branded revenue you'd have earned anyway is being counted as ad-driven. We split brand from non-brand and measure everything against your contribution margin, so the number you see maps to money you keep. We're not optimizing for a screenshot-worthy ROAS.

Do you manage Google Shopping feeds or just campaigns?

Both — in Shopping they’re inseparable. Your feed determines which auctions you can even enter, so we optimize titles, product types, attributes, and data quality alongside campaign structure. Treating the feed as someone else’s job is one of the most common reasons Shopping underperforms.

Should I run Google Ads or Meta ads first?

They do different jobs: Meta creates demand from people who've never heard of you; Google captures demand from people already searching. Most scaling brands need both, but the right starting point depends on your search volume, margins, and price point — above roughly $150, buyers research before they purchase, and Google's role gets a lot bigger. That's exactly what the free audit tells you.

What budget do I need for Google Ads to be worth it?

Enough to exit the learning phase and produce readable data for your price point — there's no universal floor. What matters more is that the math works: we model what a first-time customer costs against what they're worth over time before recommending any spend level, so the budget is derived from your economics, not a package tier.

Google ads for ecommerce brandsGoogle says your ads are printing money. Your bank account disagrees.

Bestie Media is a Google ads agency for ecommerce brands that separates real new-customer growth from branded revenue you were getting anyway.

Get your free audit →

The quiet tax on sloppy Google accounts

Here's the trick most ecommerce PPC reporting plays on you: Performance Max grabs your branded searches — people already typing your name into Google — takes credit for the sale, and hands back a gorgeous blended ROAS. Meanwhile the campaigns that were supposed to find strangers are starving, your product feed has titles written for humans instead of the query engine, and every month you pay Google a commission on customers you already owned.

The account looks efficient. The business isn't growing. That's the cost of running Google without structure, and it compounds.

For more on this trap, listen to our episode Why Your ROAS Is Holding Your Brand Back.

If this is you

You know you've got product-market fit but Google feels like a black box. You can grow, but the second you scale, profit disappears. You can't tell which campaigns bring in genuinely new customers and which just intercept people on their way to your site. If that's familiar, you're exactly who we built this for.

Before we touch an account, we run the same fit check we run on every brand: does the product solve a felt problem, does it sit at a price point — roughly $150 is the line — where buyers research before they purchase, and does the category involve real comparison shopping. That tells us whether Google is going to be a growth lever or a rounding error for you.

Who it isn't for: brands looking for a set-it-and-forget-it vendor, or anyone who wants the blended-ROAS fairy tale left intact. We're going to show you the real number.

What we actually do

What we actually do on Google

Brand vs non-brand, split and measured.

First move in almost every audit: split spend the way buyers actually search —

your brand name
the product itself
the category and long-tail versions of it
open research queries

Each bucket gets its own budget, its own bid strategy, and its own honest efficiency target. Branded traffic gets defended cheaply.

Non-brand gets judged on what it actually is — new-customer acquisition — against the contribution-margin math we build before we spend a dollar. Treating all four as one undifferentiated Google budget is how brands overspend defending their own name while starving the terms that bring in strangers.

Shopping and PMax structure that you can read.

We rebuild Performance Max and Shopping around your catalog's economics — asset groups segmented by margin and role, brand exclusions, real audience signals — so PMax hunts for new demand instead of cannibalizing what you had. Then comes the filter phase: PMax will chase clicks that look like your category but aren't your buyer, so we comb the search terms and cut the misread intent with negatives as it surfaces.

We caught one residential appliance client's obvious-looking keywords quietly filling with commercial restaurant buyers — great clicks, Google pushing budget toward them, none of it converting. That filtering discipline is most of what separates a PMax account that works from one that burns budget.

Feed optimization.

In Shopping, your product feed is your ad. We rework titles around how buyers actually search, clean up product types, attributes, and the metadata in your catalog and on your pages, and fix the data quality issues that silently cap your impression share. It's its own lever, separate from bidding or budget — the boring stuff nobody brags about is the difference between scaling cleanly and lighting money on fire.

Search intent capture, top to bottom.

"Google is not a growth channel" — you hear it constantly, and it's half right: search captures demand, it doesn't create it. The half everyone forgets is YouTube. We've run YouTube Demand Gen campaigns that reported an in-platform ROAS near 0.1 — a number that gets most campaigns paused on sight — while driving a 50-60% lift in conversions everywhere else, including PMax. Judge YouTube by its own dashboard and you'll kill a channel that's working.

The two ends belong together on considered purchases. One e-bike brand we worked with, selling $4,000-5,000 bikes, had paid search dialed — yet post-purchase surveys showed 65% of buyers discovering the category on YouTube, where the brand had no presence at all. A subscriber giveaway built the channel fast, search stayed ready to close the sale when the research ended, and sales tripled.

Search doesn't do the work of creating demand on Meta — it catches what that work creates — so we map every intent that exists for what you sell: category terms, problem-aware queries, competitor-adjacent searches, high-intent long tail, each met with the right message and landing page. And when we test a bigger Google push, we hold Meta flat for two weeks and move only the Google lever, so the result is readable instead of guessed at.

We laid out the whole playbook in Stop Sleeping on Google: How Smart Brands Balance Demand Generation and Capture.

Proof

Proof, not promises

$125M+

Platform-attributed revenue

22+ months

Average client relationship

“We have loved the strategy and expertise Bestie has brought to our brand.”

Cara Loren — CEO of Cara Loren Active

Piper + Scoot

Clients have trusted us with this across brands like BEIS, Chatbooks, PhoneSoap, Jones Sports, Brixley, and Piper + Scoot. Across accounts we've driven $125M+ in platform-attributed revenue, and our average client relationship is 22+ months — brands don't stay two years with an agency whose numbers don't hold up outside the dashboard.

Here's the pattern behind a lot of those numbers. On a recent set of accounts, non-brand Google spend was sitting at 5-10% of total budget while category and research-phase demand went unclaimed.

We moved it to 40-50% — product, category, and research terms, never brand — without cutting Meta a dollar. Meta performance improved, and those brands hit their best new-customer days of the year in October, historically one of the softest months on the calendar.

We now check for that gap on every new account. Media buying here starts with your margins, not your keywords.

What changes

Right now there's a gap between the ROAS Google reports and the profit Google produces. Close it, and everything gets calmer: you know what branded defense costs, you know your true cost to acquire a stranger, PMax works for you instead of grading its own homework, and scaling becomes a budgeting decision instead of a leap of faith.

The correction runs both ways, too. We don't take any platform's attribution at face value — short post-purchase surveys asking how long a customer was searching and where are part of how we read every account — because we've watched a brand pour spend into Google when Meta's dashboard "wasn't working," only for surveys to reveal Meta was the real discovery channel all along.

Google was just where people came back to buy. The goal isn't more Google. It's the honest number.

Sometimes the real ceiling isn't Google at all — it's what happens after the click. If your landing pages and retention are the actual bottleneck, that's a Shopify marketing engagement, not another ad tweak. And whether your channel mix eventually includes TikTok ads or stays Google-first depends on your category and margins, not a template.

The first step is free.

Request your free audit and we'll get into the weeds of your account — offers, margins, customer data, campaign structure — and show you the fastest path to growth that actually makes you money. You'll get a clear next step for your brand either way. Request your free audit and see the real number behind your account.

Get your free audit →

Frequently asked questions

Is Performance Max good or bad for ecommerce brands?: It's a powerful tool that defaults to flattering itself. Left unstructured, PMax leans on branded search and remarketing to inflate its reported return. Structured properly — brand exclusions, margin-based segmentation, clean feed data, real audience signals — it's one of the best prospecting engines Google has ever shipped. The difference is in the setup, then in the filter phase after launch: combing search terms for the intent the algorithm misread and cutting it with negative keywords as it surfaces.
Why is my Google Ads ROAS high but my profit flat?: Almost always attribution, not performance: branded revenue you'd have earned anyway is being counted as ad-driven. We split brand from non-brand and measure everything against your contribution margin, so the number you see maps to money you keep. We're not optimizing for a screenshot-worthy ROAS.
Do you manage Google Shopping feeds or just campaigns?: Both — in Shopping they’re inseparable. Your feed determines which auctions you can even enter, so we optimize titles, product types, attributes, and data quality alongside campaign structure. Treating the feed as someone else’s job is one of the most common reasons Shopping underperforms.
Should I run Google Ads or Meta ads first?: They do different jobs: Meta creates demand from people who've never heard of you; Google captures demand from people already searching. Most scaling brands need both, but the right starting point depends on your search volume, margins, and price point — above roughly $150, buyers research before they purchase, and Google's role gets a lot bigger. That's exactly what the free audit tells you.
What budget do I need for Google Ads to be worth it?: Enough to exit the learning phase and produce readable data for your price point — there's no universal floor. What matters more is that the math works: we model what a first-time customer costs against what they're worth over time before recommending any spend level, so the budget is derived from your economics, not a package tier.