Google Ads for Shopify: The Meta-First Founder's Playbook

By Arthur Falcone · Founder of Arlo

July 202614 min read
Paid Ads & Acquisition

Google Ads captures demand that already exists. Meta creates it. For a Shopify brand doing $1M-$10M a year, Google is the intent-capture layer of your paid program: start with branded search to defend your name, add Shopping or Performance Max through the Google & YouTube app, and give the whole thing 10-20% of your paid budget before you test non-branded search.

If you built your paid program on Meta, you're in good company. Most Shopify founders in this range did, because Meta is where you create demand for a product nobody is searching for yet. But once your brand has real awareness, people start typing your name and your category into Google. Every one of those searches is a purchase decision happening without you in the room, and some of them end on a competitor's ad.

This guide covers where Google fits next to Meta, which campaign types deserve your money and in what order, how to split the budget, and how to measure both platforms without letting each one take credit for the same orders.

#Table of Contents

#Why is Google Ads different from Meta for a Shopify brand?

Meta interrupts people who weren't shopping. Google answers people who were.

That single difference explains almost everything about how the two platforms behave. On Meta, you pay to put a product in front of someone scrolling past their cousin's vacation photos. The ceiling is enormous because the audience is everyone, but you carry the full weight of convincing a stranger to care. On Google, someone already typed "collagen powder unflavored" or your brand name into a search box. You're not creating the intent. You're deciding whether you or a competitor gets to answer it.

Three practical consequences follow from that:

  • Google is steadier. Search demand for your category doesn't swing the way Meta performance does when a creative fatigues or an algorithm update reshuffles delivery. Campaigns that work tend to keep working with less babysitting.
  • Google has a lower volume ceiling. You can't spend your way past the number of people searching for your category this month. Meta scales with budget and creative. Google scales with demand that already exists, and it flattens when you've captured it.
  • Google defends what Meta builds. Some meaningful share of the people who see your Meta ads don't click. They search your brand name two days later. If a competitor is bidding on your name, part of your Meta budget is quietly funding their pipeline.

This is why "Meta versus Google" is the wrong frame. They do different jobs. Meta fills the top of the funnel; Google catches what falls toward the bottom. If you're still deciding how paid channels fit together across your whole business, the broader framework in our guide to advertising in e-commerce is the place to start. This post assumes you've made that call and Meta is already running.

One expectation to set before you open the account: your first Google campaigns will post ROAS numbers that look spectacular next to Meta. That's not because Google is a better platform. It's because branded search and Shopping harvest intent that your brand, your organic rankings, and your Meta spend already created. Keep that in mind every time you're tempted to compare the two platforms on their in-platform numbers.

#The campaign stack for a $1M-$10M Shopify brand

Google Ads offers more campaign types than you need: Search, Performance Max, standard Shopping, Demand Gen, Display, App, and video campaigns on YouTube. For a Shopify brand at this stage, only four matter, and they matter in a specific order.

#1. Branded search: defend your name

A branded search campaign bids on your own brand name and product names, usually in exact and phrase match. It's the cheapest, fastest campaign you will ever launch, and across the Shopify stores Arlo watches, it's routinely the highest-ROAS line item in the entire ad account. That's exactly why you should be suspicious of it.

Branded clicks are cheap because Google rewards relevance, and nothing is more relevant to a brand query than the brand. The real question is incrementality: would those people have bought anyway through your organic listing?

Branded search is genuinely worth paying for when:

  • Competitors bid on your name. Search your brand in an incognito window. If a rival's ad sits above your organic result, every branded click you don't buy is a coin flip you might lose.
  • Shopping ads appear on your brand query. Product listing ads with images and prices push organic results down the page, and those units can belong to resellers or competitors.
  • You're spending heavily on Meta or TikTok. Demand you paid to create comes back through search. Losing it at the last step is the most expensive way to lose it.

Branded search is mostly wasted when nobody bids against you and your organic listing owns the whole first screen. Don't guess. Run the test: pause branded for two weeks, then compare total orders from branded-term traffic (paid plus organic) against the prior two weeks in Shopify. If total branded orders hold while paid spend goes to zero, you were buying your own traffic. Across the stores Arlo sees at this revenue range, a defensible branded campaign rarely needs more than $500 to $1,500 a month. If yours wants more, competitors are attacking your name aggressively, and that's worth knowing too.

#2. Shopping and Performance Max: put your feed to work

Shopping ads are the image-price-title product units at the top of commercial searches, and they're where most of a Shopify brand's non-branded Google revenue actually comes from. They run off a product feed, not keywords, so setup starts with plumbing.

The plumbing is straightforward on Shopify. Install the free Google & YouTube app from the Shopify App Store, connect your Google account, and the app creates or links a Google Merchant Center account and syncs your catalog to it automatically. Product changes flow through on an ongoing basis, typically within a day or two, so prices and inventory stay accurate without manual feed exports. Once the feed is live, spend an hour in Merchant Center clearing disapprovals. Missing GTINs, mismatched prices, and unconfigured shipping settings are the usual offenders, and every disapproved product is a product Google won't advertise. A side benefit: an approved feed also gets you free listings in Google's Shopping tab, which costs nothing and occasionally surprises you.

Feed quality is the highest-leverage work in this whole section. Google matches Shopping ads to queries using your product titles and attributes, so "Lumira Renewal Serum" gets found by almost nobody, while "Lumira Vitamin C Face Serum for Dark Spots, 1 oz" gets found by people typing exactly what they want.

With the feed live, you have two ways to run it:

  • Standard Shopping gives you control: full search-term visibility, negative keywords, and manual bidding. It's the right starting point when your Google budget is small and you want to see exactly which queries you're paying for.
  • Performance Max is Google's automated campaign type. One campaign serves your feed and creative assets across Search, Shopping, YouTube, Display, Discover, Gmail, and Maps, with Google's machine deciding where each dollar goes. For most e-commerce accounts it finds more volume than standard Shopping, and Google has steered retail advertisers toward it hard.

PMax deserves two cautions. First, it will cannibalize your branded traffic by default. Left alone, PMax serves on your own brand queries, hoovers up those cheap high-converting clicks, and reports them as campaign performance. The fix is to add your own brand to a brand exclusion list on the PMax campaign, and to use campaign-level negative keywords, which Google finally opened to all PMax advertisers in 2025. That keeps branded traffic in your branded campaign where you can see and price it honestly. Second, PMax is still a partial black box. Google has expanded channel-level and search-term reporting over the past two years, but you'll never get the query-by-query control of a standard Search campaign. You're trading transparency for reach, and the trade is only worth it when you've protected your brand terms first.

A practical rule: under roughly $3,000 a month of Shopping budget, start with standard Shopping. Above that, run PMax with brand exclusions in place from day one.

#3. Non-branded search: buy intent you don't own yet

Non-branded search means bidding on category queries: "running hat for small heads," "sulfate free dog shampoo," whatever your customers type before they know you exist. This is the most competitive, most expensive corner of Google, and it's where unit economics decide everything.

The math is the same breakeven discipline you should already be running on Meta. If your contribution margin after COGS, shipping, and payment fees is 35%, your breakeven ROAS is 1 ÷ 0.35, or about 2.9. Non-branded search has to clear that bar on honestly measured revenue, with no help from branded traffic mixed into the same campaign. The full breakdown of why margin sets your ROAS floor is in our guide to ROAS vs ROI.

Across Shopify stores Arlo watches, non-branded search tends to work when at least one of these is true: AOV is high enough that a $2-4 click doesn't wreck the order math, the product has strong repeat purchase so first-order breakeven is acceptable, or the query is specific enough ("gluten free protein powder chocolate") that conversion rates run well above sitewide average. It tends to fail on low-AOV, one-and-done products in broad categories, where you're bidding against Amazon and brands with deeper pockets.

Start narrow: exact-match keywords on your hero product's most specific queries, one theme per ad group, and a landing page that matches the query rather than your homepage. Expand only after the narrow version clears breakeven for a month.

#4. YouTube and Demand Gen: later, if at all

Demand Gen is Google's demand-creation campaign type, serving video and image ads across YouTube (including Shorts), Discover, Gmail, and the Display Network. Google folded its old Video Action Campaigns into Demand Gen, so this is now the main path to YouTube performance buying.

Notice what job that is: creating demand from people who weren't shopping. That's Meta's job, and Meta is already doing it for you with a creative pipeline you've spent years tuning. Demand Gen makes sense when branded and Shopping are saturated, non-branded is at breakeven or better, and you have video creative worth showing. For most brands in this range, that's a next-year conversation, not a this-quarter one. Skipping YouTube entirely while you build out steps one through three costs you almost nothing.

#How much should you spend, and how do you phase it in?

The heuristic across Meta-first Shopify brands Arlo watches: Google starts at 10-20% of total paid budget and, where non-branded earns its keep, settles around 25-35%. Meta stays the majority channel because it's the engine that creates demand; Google's share is capped by how much intent exists to capture.

Here's what phasing in looks like for a concrete case. Say you're doing $250,000 a month in revenue and spending $50,000 a month on paid, nearly all Meta.

Months 1-2: carve out $5,000 a month. Put roughly $1,000 into branded search and $4,000 into Shopping (standard Shopping, or PMax with your brand excluded). Don't fund this by gutting a proven Meta campaign; trim your weakest Meta ad sets, the ones you were half-planning to kill anyway. Spend the first two weeks fixing feed disapprovals and mining the search terms report for negatives.

Month 3: judge it on blended numbers. If your overall MER held steady or improved while Google took over some volume, the channel is earning its slot. If Google's in-platform ROAS looks great but blended MER didn't move, Google is mostly re-labeling orders you were already getting, usually via branded cannibalization. Check the brand exclusions before you check anything else.

Months 3-6: scale what cleared the bar. Push Shopping/PMax toward $8,000-10,000 and, if your margin math supports it, test $1,500-2,000 on narrow non-branded exact match. Keep branded flat; it doesn't need to grow just because everything else did.

The discipline that matters most: scale Google in steps of 20-30% and watch blended MER for two weeks after each step. Google's steadiness makes it tempting to scale fast, but the intent pool is finite, and the first sign you've outrun it is a rising cost per conversion on the same query set.

#How do I measure Google Ads alongside Meta?

The week you turn Google on, you'll notice something uncomfortable: Meta and Google will both claim some of the same orders, at full value, in the same week.

A customer sees your Meta ad on Tuesday, searches your brand on Thursday, clicks your branded search ad, and buys. Meta counts that order under its 7-day click attribution. Google counts it too, because the last click was Google's. Both platforms are telling the truth by their own rules, and adding their numbers together now overstates reality. In-platform figures were already optimistic before this, for reasons we've covered in why Meta Ads ROAS is broken, and multi-platform double-counting stacks on top. If you want to understand the mechanics of who claims what, our breakdown of multi-touch attribution models covers it.

The tiebreaker is blended MER: total store revenue divided by total ad spend across every platform, tracked weekly. MER can't be gamed by attribution windows because it ignores them. The full comparison of when to use each metric is in MER vs ROAS vs ROI, but the operating rule is short:

  • Use platform ROAS to compare campaigns within a platform. Which Shopping campaign gets more budget, which ad set gets killed. The numbers are inflated, but they're inflated consistently, so relative comparisons hold.
  • Use blended MER to judge the addition of a channel. You added $5,000 of Google spend. Did total revenue rise enough to hold your MER? That's the whole test, and neither platform's dashboard can answer it.

One Google-specific trap: because branded search rides on demand the rest of your marketing created, lumping it into "Google performance" flatters the channel. When you review Google weekly, look at branded and non-branded as separate lines. Branded answers "are we defending the name efficiently?" Non-branded and Shopping answer "is Google actually acquiring customers?" Those are different questions, and one blended Google ROAS number hides both answers.

#When should you kill a Google Ads campaign?

The same breakeven-ROAS discipline you run on Meta applies here, with one adjustment for how Google fails.

Meta campaigns usually fail loudly. Creative fatigues, CPA spikes, and the decision announces itself inside a week. Google campaigns fail quietly. A Shopping campaign drifts onto irrelevant queries, a PMax campaign leans harder into remarketing and brand-adjacent traffic, and spend keeps flowing at a slowly decaying return that never produces a single alarming day. Nobody kills it because it never has a bad enough week to trigger the conversation.

So put the check on a schedule instead of waiting for an alarm:

  • Know your breakeven ROAS (1 ÷ contribution margin) and hold every non-branded campaign to it. A campaign below breakeven for 3-4 consecutive weeks with real volume behind it gets fixed or killed, the same rule laid out in when to kill a losing ad.
  • Read the search terms report weekly. Google campaigns rarely die from bad ads. They die from query drift, and negatives are the fix long before the topline shows the damage.
  • Give Google longer windows than Meta. Lower volume means noisier weekly reads. Judge non-branded search on 2-4 week blocks, not single days, but don't let "it needs more data" stretch past a month of below-breakeven spend.
  • Never judge branded on ROAS. Its ROAS always looks incredible. Judge it on incrementality with the pause test from earlier, once or twice a year.

If you're staring at a specific campaign right now trying to make the call, run the numbers through our free calculator: Should I kill this ad? It applies the breakeven math for you.

#FAQ

#Are Google Ads worth it for a Shopify store?

Yes, with the right expectations. Google Ads captures existing demand, so it delivers steadier but lower-volume results than Meta. For most Shopify stores, branded search and Shopping campaigns pay for themselves quickly because they reach shoppers already looking for you or your product category. Treat Google as a complement to Meta rather than a replacement, and size it at 10-20% of your paid budget to start.

#How much should a Shopify store spend on Google Ads?

Start with 10-20% of your total paid budget. A brand spending $50,000 a month on ads might begin with $5,000 on Google: roughly $1,000 on branded search and $4,000 on Shopping or Performance Max. Hold that level for 60 to 90 days, judge it by blended MER rather than platform ROAS, then scale toward 25-35% of paid spend if the economics hold.

#Should I use Performance Max or standard Shopping campaigns?

Standard Shopping gives you more control: full search-term visibility, negative keywords, and manual bids. Performance Max usually finds more volume because it serves across Search, Shopping, YouTube, Display, Discover, Gmail, and Maps. If your Google budget is under about $3,000 a month, start with standard Shopping. Above that, run Performance Max with your own brand excluded so it cannot take credit for branded traffic.

#Should I bid on my own brand name?

Bid on your brand if competitors show up when you search it, or if Shopping ads appear above your organic listing. Those clicks are cheap insurance. If nobody bids against you and you own the entire first screen organically, branded ads mostly buy traffic you would have received free. Test it: pause branded for two weeks and watch whether total branded orders in Shopify actually drop.

#Do I need the Google & YouTube app to run Shopping ads?

You need a Google Merchant Center account with an approved product feed, and the free Google & YouTube app in the Shopify App Store is the simplest way to get one. It connects your store to Merchant Center and syncs your catalog automatically, so price and inventory changes flow through without manual feed maintenance. Larger catalogs sometimes graduate to dedicated feed apps for finer control over titles and attributes.

#Watch Meta and Google in one place

The hard part of running both platforms isn't setup. It's the weekly judgment call: two dashboards, both claiming credit, both asking for more budget. Arlo watches Meta and Google side by side against your real Shopify revenue and margins, then tells you each week where the next dollar should go, in plain English. It's $47/month with a 14-day free trial, and it takes about two minutes to connect your store.

Your weekly marketing direction, built from your Shopify data.

Free for 14 days. Then $47/month.