Ecommerce Google Ads: How I Run Shopping, Performance Max and Search

Last updated: 7 November 2026

Ecommerce Google Ads work when Shopping, Performance Max and Search each have one clear job and clean data behind them. I have managed paid search since 2013, and this is how I set up Google Ads for ecommerce stores, with real numbers from a Judaica store and P-Rex Hobby. It walks through the structure and the reasons behind it, not a list of tips.

Quick Answer: How should an online store set up ecommerce Google Ads?

Fix conversion tracking first. Then run Shopping or Performance Max from a clean product feed, and Search on exact match for the product terms you most want to own. Add account-level negatives before launch, keep Performance Max off non-selling pages, and move to target ROAS only once results hold steady above your break-even ROAS.

Here is how each campaign type splits the work in the setup I use.

Which Google Ads campaigns does an online store need?

Most stores need three working together: Shopping or Performance Max for product listings, and Search for the terms you most want to own. Display and video can come later, once the core is profitable.

Google Shopping ads for ecommerce have no keywords. Google matches each search to your product data from Merchant Center, so the feed does the targeting. Performance Max can use the same feed and reaches YouTube, Display, Search, Discover, Gmail and Maps from one campaign, according to Google’s Performance Max overview.

Search fills a different gap. It lets you choose the exact queries, write the ad yourself and send each click to the best page. For a store, that control matters most on high-margin products, key categories and your own brand name.

Can Shopping, Performance Max and Search run in the same account?

Yes, but they compete for some of the same searches, so you need to know the rules. Google says that when campaigns could serve the same impression, the one with the highest Ad Rank serves. A Search keyword that exactly matches the query is the exception, because it is preferred over Performance Max.

Google adds that Performance Max may filter out brand and navigational searches so dedicated Search campaigns can handle them. Both points come from its page on how Performance Max interacts with other campaigns. The practical upshot is simple: put your most important terms in Search on exact match, and let PMax pick up the rest.

Why do I check conversion tracking before touching bids?

Because Smart Bidding learns from the conversions and values you report. Google’s target ROAS page says the strategy predicts future conversions and values from your reported conversion values. If those values are wrong, every bid built on them is wrong too.

So before I change bids or structure, I check GA4, Google Tag Manager and enhanced conversions. Enhanced conversions send hashed first-party data, such as an email address, to improve the accuracy of conversion measurement, according to Google’s enhanced conversions page.

The checks I run on a store account look like this:

1. Count each purchase once, with a single primary purchase action feeding bids.
2. Pass the real order value and currency with every purchase.
3. Confirm in Google Tag Manager that the tag fires once per order, not on every page reload.
4. Turn on enhanced conversions where the store collects an email at checkout.
5. Match a recent order across Google Ads, GA4 and the store backend.

Only when those line up do I touch structure or bidding. A ROAS that looks great on double-counted orders is worse than no data, because it teaches the system the wrong lesson.

How do Search and Performance Max share the work?

Search on exact match owns the terms you care about most, and Performance Max covers the wider demand once there is data. That split keeps traffic relevant early and gives broad reach to the campaign built for it.

Why do I start new Search campaigns on exact match?

Exact match costs more per click, but the traffic matches the products you sell. I run new Search campaigns on exact for roughly the first 15 days to build relevant audience and conversion data. Broad match only works once the account already has conversion data.

On low-volume accounts that stay exact-only, cost per click can run high. There I test manual CPC to control what each click costs. Accounts with enough conversion data go on Maximize conversion value, and later target ROAS.

When do I let Performance Max do the broad job?

Once the account has conversion data. Instead of opening Search to broad match early, I let Performance Max find the wider searches. It has the feed, the assets and the conversion history to work with, and Google positions it as a complement to keyword-based Search campaigns.

What guardrails do I set before launching Performance Max?

Three: account-level negative keywords in phrase and exact match, exclusions for irrelevant placements, and URL exclusions for blog posts and other pages that do not sell. With those in place, PMax sends paid traffic only to pages that can convert.

My pre-launch checklist for a store:

• Account-level negatives in phrase and exact match, built before launch rather than after waste appears
• Placement exclusions for sites and apps that do not fit the brand
• URL exclusions for the blog, careers, policy pages and other non-selling URLs
• Conversion goals set to purchases, with real order values
• A feed check for disapprovals, stale prices and out-of-stock items

Google’s documentation backs each piece. Account-level negatives apply to Search and Shopping inventory in Performance Max, per its PMax negative keywords page. The account-level list holds up to 1,000 terms.

URL exclusion keeps chosen pages out of PMax ads, though Google notes a Merchant Center campaign’s homepage may still serve on some inventory. Account-level placement exclusions also apply to PMax, according to Google’s brand suitability guide.

Why account level and not only campaign level? Because a negative you miss in one campaign stops mattering when the account-level list already blocks it. One list, reviewed against the search terms report, protects every campaign at once. My Performance Max campaign guide covers the full setup.

How does the product feed decide where Shopping ads show?

Google reads your titles, descriptions and attributes to decide which searches a product can match. A vague title gets vague traffic, while a title that names what buyers type gets the searches you want.

Google’s title guidance allows up to 150 characters, asks for the most important details first and suggests using the keywords buyers search for. It also notes that shoppers usually notice only the first 70 or fewer characters.

I pick title and description keywords from real search demand: Keyword Planner, Semrush volume and the search terms report. I do not let AI rewrite titles on a guess, because a nicer-sounding word that nobody searches for can cost you the match. My Google Shopping feed optimization guide walks through the method.

Price and stock matter as much as words. When a product sells out or a price changes, the feed has to change with it. Otherwise paid clicks land on a page that cannot take the order.

What happened in the Judaica store account?

Between two 120 day windows, Google Ads spend fell about 37% while conversion value rose about 120%. ROAS went from 4.90 to 17.15, as the Judaica ecommerce case study shows.

The store is a long-established Judaica and Jewish books retailer with showrooms and a national online store. Its catalog runs from books and holiday items to gifts and children’s products. My team handles its Google Ads and Meta Ads, along with SEO and the product feed.

Here are the two windows side by side. The earlier figures are derived from interface deltas, as the case study notes.

Source: Judaica ecommerce Google Ads and Meta Ads case study, Hustle Marketers. Platform-reported figures, read from the live account on 2 October 2026.

What changed in the account?

Budget moved from weaker campaigns to the ones that match named-product demand. The case study’s reading is that the old account was buying the wrong traffic, not expensive traffic, since cheap clicks were already available.

The shift was not even. Of four Google campaigns, two grew conversions and two went backwards, while the account as a whole more than doubled conversion value. The largest contributor added 274.95 conversions, per the case study.

Why was the product feed the main lever?

Because remarketing was off the table. Google treats religious products as a sensitive interest category, so advertiser-curated audiences such as Customer Match and your data segments cannot be used.

With no lists to lean on, accurate titles, availability and categorization decide what gets shown. The same applies to other restricted categories, such as alcohol: when audiences are limited, the feed does the targeting.

How did Meta Ads do on the same store?

Meta ran alongside Google. Across the four largest of eight Meta campaigns, blended purchase ROAS was 15.47 between 1 June and 30 September 2026. All four ran on Highest volume bidding rather than a ROAS target, per the case study.

Some caveats apply to both platforms. Each one credits sales with its own attribution, the figures are not deduplicated, and the two Google windows are not a controlled test. ROAS here is revenue against ad cost, not profit.

What did P-Rex Hobby’s account show?

P-Rex Hobby is an anime figure and collectibles store. Its account data shows a month at 1000% ROAS, which means about $10 in revenue for every $1 of ad spend, with $98,000 in revenue that month. It is one month in one niche, so treat it as a data point, not a forecast for your store.

P-Rex Hobby shares its side of the work in this short video.

What do other ecommerce owners say?

Here is another store owner’s view. A German ecommerce brand owner describes what working with me on Google Ads was like.

When do I move an ecommerce account to target ROAS?

After ROAS has held steady for about 30 days on Maximize conversion value with no target. Then I switch to target ROAS and push the target up by 15% to 20%. My target ROAS guide covers each step.

Google’s target ROAS page explains that the strategy predicts conversion value and sets bids to maximize it. It asks for at least 15 conversions in the past 30 days on Search and Shopping. It also warns that a target set too high can limit traffic.

The floor for any target is break-even ROAS, which is 1 divided by your profit margin. The table shows the arithmetic for a few margins.

Source: arithmetic only, using break-even ROAS equals 1 divided by profit margin. These are example margins, not client figures.

I never promise a ROAS before seeing an account, especially for a relaunched campaign still in learning. What I commit to is the process: clean tracking, a tight structure and every decision measured against break-even.

How do you win new customers without resetting live campaigns?

Leave the campaigns that already work as they are, and test new-customer acquisition in a separate new campaign that excludes your existing customer list. Changing goals on a live campaign can put it back into learning, and that is a cost you do not need.

Google’s customer lifecycle goals include two modes that work in both Shopping and Performance Max. New Customer Value bids higher for new customers than existing ones, and New Customer Only bids only for new customers. Google asks you to define existing customers with first-party data, such as a customer list and a website tag.

Running the test in its own campaign keeps the comparison clean. If it works, scale it. If it does not, the campaigns paying the bills never felt the experiment.

FAQs

Is Google Ads worth it for a small ecommerce store?

It can be, if tracking works, the feed is clean and your margins leave room above break-even ROAS. A store with thin margins needs a very high ROAS just to break even, so I check the numbers before suggesting any spend. I never promise a ROAS, because results depend on the products, prices and competition.

Should I start with Shopping or Performance Max?

In my setup, Search on exact match comes first, for the best product terms, with the product feed cleaned up from day one. Performance Max takes the broad job once the account has conversion data. The order matters less than having tracking, negatives and URL exclusions in place before anything spends.

Do I need Google Merchant Center for ecommerce Google Ads?

You need it for Shopping ads and for Performance Max campaigns that show your products, because both read product data from Merchant Center. Search campaigns can run without it. In practice almost every store benefits from a feed, since product listings show the image and price before anyone clicks.

Should I add negative keywords to Performance Max?

Yes. I build an account-level list in phrase and exact match before launch, so it applies to Search, Shopping and the search inventory inside Performance Max. Google says these negatives cover Search and Shopping inventory only, so I also exclude irrelevant placements and non-selling URLs to keep the rest of the campaign on track.

What is a good ROAS for ecommerce Google Ads?

Any ROAS above your break-even point, which is 1 divided by your profit margin. A store that keeps half of each sale as profit breaks even at 2. Results vary widely by margin, price and competition, so I judge campaigns against each store’s own break-even figure instead of an industry average.

How much does a freelance Google Ads specialist cost for an online store?

My fee is $600 to $800 per platform a month, month to month, with no lock-in. You own every account, and ad spend is paid to Google separately. White label pricing for agencies is custom. You can see how I work on my freelance Google Ads specialist page.

Can I run remarketing for every product category?

No. Some categories are sensitive under Google’s personalized advertising rules, including religious products and alcohol, and they cannot use advertiser-curated audiences such as remarketing lists. In those accounts the product feed, the keywords and predefined Google audiences carry the targeting, which is what I saw on the Judaica store.

What should you do next with your ecommerce Google Ads?

Start with tracking, because every other decision depends on it. Then check that Search, Shopping and Performance Max each have one job, that one account-level negative list covers them all, and that PMax cannot spend on pages that do not sell.

I have managed paid search since 2013 and founded Hustle Marketers in 2020. Client accounts my team and I worked on have generated $780M+ in trackable revenue, with 591+ client reviews across platforms, as listed on my Hustle Marketers profile.

On Upwork I hold Top Rated Plus, which Upwork describes as the top 3% of talent. You can hire a freelance Google Ads specialist for your store, or read more guides on the Google Ads blog.

How this post was written: drawn from Ishant Sharma’s own account work and the published case studies linked above, researched against current Google Ads documentation, drafted with AI assistance and checked against the sources before publishing.