Google Shopping Ads: Product Listing Ads and Shopping Campaigns, Run by AI

Google Shopping ads cost an average of about $0.66 per click in the US, with most retail categories landing between roughly $0.46 and $1.20 and competitive categories running several dollars. There is no minimum spend and no minimum bid, so the real cost is set by your daily budget rather than by Google. Published benchmarks put the average Shopping cost per acquisition near $39 at a conversion rate near 1.9 percent.

Those averages are only useful as a sanity check. What decides whether Shopping is affordable for you is your gross margin per order, not the click price. A store with a $90 average order value and a 40 percent margin has about $36 of gross profit to work with, so a $39 acquisition cost is a loss even though it matches the benchmark exactly. The same $39 against a $400 order value is excellent. Run your own numbers through the ROAS calculator before you judge any benchmark.

On top of the ad spend there is the cost of someone managing it. That is typically 10 to 20 percent of ad spend with an agency, $75 to $200 an hour with a freelancer, or a subscription with a software tool. AdBot is a flat $297 to $1,497 a month regardless of budget. If you are still sizing the ad budget itself, the advertising budget calculator works backwards from your target number of customers.

Neither is better in general, and most healthy ecommerce accounts run both. Standard Shopping gives you product-level control and a full search terms report. Performance Max reaches further because it can also serve on YouTube, Display, Gmail and Maps, but it decides most of the targeting itself and reports back in themes rather than exact queries. The right split depends on how much conversion data your account generates each month.

A practical rule that holds up well: if your account produces fewer than about 30 conversions a month, Performance Max does not have enough signal to optimize with and standard Shopping usually wins, because you can steer it manually. Above that threshold Performance Max often finds incremental demand that Shopping alone misses. Many accounts end up running standard Shopping on their best-margin products at a high campaign priority and Performance Max across everything else.

The complaint you will read most often about Performance Max is not that it performs badly, it is that it is hard to see inside. That is a real cost when something breaks, and it is why we keep standard Shopping in the mix rather than moving whole catalogs into a single automated campaign. There is more detail on getting the automated side under control on our Performance Max optimization page.

No. Google Shopping ads do not use keywords. Google decides which queries your products can appear for by reading your product feed, so your titles, descriptions, product types, brand fields and attributes are what determine your reach. You cannot add a keyword to a Shopping campaign, but you can add negative keywords, which is how you stop the campaign appearing on queries that will never convert.

This is why feed quality is the whole game in Shopping. Put the words a buyer would actually type into the front of the product title: brand, product type, then the attributes that distinguish it, such as size, color, material or model. Google truncates titles in the display, so the first 60 to 70 characters carry most of the weight. Vague or internal naming conventions are the single most common reason a well-funded Shopping campaign gets almost no impressions.

You need four things: a Google Merchant Center account, a product feed that passes Merchant Center validation, a linked Google Ads account, and a website that meets Google's retail policies. The policy requirements are the part that catches people out, so it is worth checking them before you build anything.

Yes, and often better than paid social does, because a small store does not have to create demand to use Shopping. Somebody is already searching for the product. The constraint is not store size, it is whether you have enough margin per order to absorb an acquisition cost in the $20 to $50 range, and whether you can generate enough monthly conversions for the bidding to stabilize.

Where small stores struggle is competing on identical products against large retailers who can outbid them all day. If you resell the same SKUs as Amazon and three national chains, Shopping will be brutal on your generic terms. The way out is to weight your budget toward the products where you are genuinely differentiated: exclusive lines, bundles, unusual sizes, your own brand, or categories the big retailers cover badly. That is a product-group and budget decision rather than a bidding trick.

If you sell on Shopify, the mechanics of the feed and the Merchant Center link are mostly handled for you, which removes a real chunk of the setup work. We cover that path specifically on the Google Ads for Shopify page, and the broader automation picture on AI ads for Shopify.

Start with the number of conversions you need rather than a budget figure. Google's bidding models need roughly 30 conversions in a 30 day window before they optimize reliably. At an average Shopping cost per acquisition near $39, that implies somewhere around $1,200 a month as a working floor for a single campaign, though a store with a low average order value and a high conversion rate can get there for less.

Below that floor the honest advice is to narrow rather than to spread. Run Shopping on your best twenty products instead of your whole catalog, get the data to stabilize, then widen. Splitting $600 a month across 2,000 SKUs produces a trickle of clicks on everything and enough signal on nothing.

Once the campaign is stable, the useful question changes from how much to spend to where the next dollar goes. That is a cross-channel decision, and it is the one most vendors avoid because they only run one channel. Our ecommerce PPC services page covers how Shopping, Search and paid social should share a single budget.

There are four realistic options and they suit very different situations. Doing it yourself is genuinely viable under about $2,000 a month in spend if you can give it two focused hours a week. A freelancer at $75 to $200 an hour is good for a rebuild but rarely gives you daily attention. An agency at 10 to 20 percent of spend brings a team, and makes most sense above roughly $15,000 a month when that percentage buys real seniority.

The fourth option is software that does the work rather than advising you about it. That is where AdBot sits. It suits a store that has products worth advertising, no in-house paid media person, and a budget where a percentage fee would buy a junior account manager and a monthly check-in. Where it does not suit you: if you already employ a strong ecommerce PPC specialist, tools that multiply that person are a better buy, and we say so on our best AI ad tools roundup.

Whatever you choose, insist that campaigns are built inside a Merchant Center and Google Ads account registered to your business. Conversion history, audience lists and campaign structure are assets that take months to rebuild, and the time to establish who owns them is before the work starts, not when the relationship ends. If you are weighing the agency route specifically, our PPC agency comparison lays out the fee models side by side.

AdBot builds the Shopping campaign structure, creates product groups that reflect your margins, sets and adjusts bids, moves budget between products and campaigns on measured performance, mines the search terms report and adds negative keywords, and decides where standard Shopping and Performance Max should each own part of the catalog. It reports what it changed in plain English, and it runs inside your own accounts.

What it does not do, stated plainly so nobody buys the wrong thing: it does not build or repair your product feed, it does not fix Merchant Center disapprovals for you, and it does not do inventory-driven catalog automation such as automatically pausing ads when a variant sells out or writing live prices onto ad images. Those are feed-tool jobs. It also does not run Microsoft Shopping, Amazon Ads or marketplace listings. It covers Google Search, Shopping and Performance Max, plus Meta and TikTok.

For the rest of the Google Ads account, including Search and the account-wide automation, see automated Google Ads management. If you want a read on what is wrong with an existing Shopping setup before changing anything, start with a Google Ads audit.