In ecommerce Google Ads, the lever isn’t the bid strategy or the creative. It’s the product feed, and most of your budget is already riding on it whether you’ve optimized it or not.
A shopper who searches for a product is ready to buy. They compare a few options, read a few reviews, and order from the listing that shows the clearest title, the cleanest image, and a price that makes sense. Almost all of that decision happens inside the ad unit, before they ever reach your site. The campaign that wins is the one Google can read well enough to match to the query, and that match is governed by your feed, not your settings.
That is why a generic “run some Google Ads” approach underperforms for online stores. Performance Max already controls the majority of Shopping budget, and roughly 90% of that spend flows through feed-based product ads. So when the feed is thin (vague titles, missing attributes, weak images), you are not running a worse version of the same campaign; you are pointing real money at queries you will never win. We build around the feed first, then layer bidding, structure, and seasonality on top. Every number on this page is backed by a real source, listed at the bottom.
The case for doing this differently is not our opinion. It is what the data says, every figure sourced below.
Almost all of your Shopping budget rides on the feed, not the settings.
Smarter Ecommerce analyzed more than 4,000 Performance Max campaigns and found that between 74% and 97% of PMax spend flows through feed-based product ads, with a stable median of 90%. In other words, the product feed is not one input among many; it is where nearly all the money goes. Yet most advertisers spend their optimization time inside campaign settings and audience signals, the slice of spend that moves the least.
This is the entire thesis of feed-first ecommerce advertising. When Google can read a clean, complete, well-titled feed, it matches your products to the right queries and bids them accordingly. When it can’t, budget leaks toward broad, low-intent matches you were never going to convert. We treat the feed as the primary asset: titles, attributes, images, GTINs, and product types built to be matched, not just uploaded.
Roughly 90% of Performance Max spend rides on the feed. Optimize the feed and you optimize the budget.
Where Performance Max spends
The feed governs the budget. Range across campaigns runs 74% to 97%.
Source: Smarter Ecommerce, State of Performance Max 2025Performance Max now controls most of the money and the revenue.
This is no longer a niche format. In Tinuiti’s Q2 2025 benchmark, Performance Max made up 59% of all Google Shopping spend and drove 59% of the revenue from Shopping listings, up from 53% of spend a quarter earlier. The budget has moved into a campaign type that runs almost entirely on the feed, which is exactly why feed quality is now the controlling variable for an online store’s paid performance.
The window is favorable for disciplined buyers. In the same quarter, Shopping spend grew 19% year over year and clicks grew 18%, while cost per click rose just 1%. Demand and inventory are expanding faster than prices, so the stores that win are the ones converting that cheap, growing click volume into orders, not the ones simply bidding harder. We run Performance Max as a feed program first, with the structure and exclusions that keep it from drifting into brand traffic you would have captured for free.
Performance Max share of Google Shopping spend
A better product title is the cheapest growth lever you have.
The feed-first claim is not theoretical. Store Growers ran original research across 151 Shopping ads and documented optimized product titles lifting impressions 39% and clicks 38% within a single week, with click-through rate holding flat. No new budget, no new creative, no bid changes: the same products, described in a way Google could match to more of the right searches. That is the mechanism behind feed-first, in one experiment.
Titles are where most stores leak the easiest gains. Manufacturer-default names bury the attributes shoppers search by (brand, model, size, color, material, use case), so the listing never enters auctions it would win. We rebuild titles to lead with the terms that pull qualified queries, then extend the same discipline to product types, attributes, and images. As Store Growers’ Dennis Moons puts it, “Feed quality is everything. A product with a blurry image, a vague title, or a missing price will barely show.”
What optimized titles did in one week
Shopping clicks are cheap; conversion discipline is where the margin lives.
The Shopping math is forgiving on the front end and unforgiving on the back. The average Shopping click runs about $0.66 and converts at roughly 1.91%, with a cost per action near $38.87, lower than the $45.27 typical of standard Search ads. Cheap traffic and a sub-2% conversion rate means the economics are decided after the click: the listings that win are the ones with the right title, image, price, and review proof to convert a visitor who is comparing three tabs.
So spending more is rarely the strategy. The edge is converting the click you already paid for, which is why we pair feed work with conversion-rate work on the product and cart pages, and with the review profile that closes the sale. It compounds at scale: Google’s own data shows advertisers who adopt Performance Max see an average 27% lift in conversions or value at a similar CPA or ROAS. We report on revenue and return on ad spend, not on traffic that never checks out.
A $0.66 click that converts at under 2% means the order is won after the click, not before it.
The numbers behind a Shopping click
Ecommerce demand isn’t flat, and your budget shouldn’t be either.
Search demand for products swings hard into the holiday peak. Tinuiti’s 2024 Cyber Five data shows median retailer sales growth from Google search ads climbing from 11% year over year in October to 20% over the seven days before Thanksgiving, then spiking to 29% on Thanksgiving Day itself. A budget that holds steady through that curve underspends the moment that matters most and overspends the lulls around it.
Discovery is search-anchored and multi-channel, which raises the stakes on getting the peak right. We plan budget, feed priorities, and promotional pricing against the calendar, so the feed is clean and the bids are aggressive when demand is real, and disciplined when it isn’t.
Search ad sales growth into the peak
The same feed that wins the auction wins the AI answer.
Search is shifting under every online store. Pew Research found that 18% of Google searches now return an AI summary, and when one appears people click a traditional result far less often: 8% of the time versus 15% with no summary, and just 1% click a source cited inside the AI answer. BrightEdge already sees AI Overviews on 23% of ecommerce queries, so this is a present-tense problem for product discovery, not a future one.
The flip side is real upside. Adobe found traffic to US retail sites from generative AI sources rose 1,200% year over year during the 2024 holiday season, and those AI-referred visitors browse 12% more pages with a 23% lower bounce rate. The structured, well-described product data that makes your feed legible to Google’s ad systems is the same data that makes your catalog legible to AI answers. We build the feed and the product pages to be matched, quoted, and cited across both, so paid and AI discovery pull in the same direction.
AI answers are compressing the click
And only 1% of searchers click a source cited inside the AI summary.
Source: Pew Research Center, 2025Most advertisers optimizing Performance Max are focused on campaign settings, but the bigger opportunity is usually in the product feed.
Tony Adam, CEO at Visible Factors, in Search Engine Journal
Without an optimized product feed, Google won’t know which queries to show your products for.
Menachem Ani, founder of JXT Group, in Search Engine Land
Feed quality is everything. A product with a blurry image, a vague title, or a missing price will barely show.
Dennis Moons, Founder and Lead Instructor at Store Growers
Ready to turn Google Ads into a revenue line, not a traffic bill?
If your Shopping and Performance Max campaigns are running on a feed nobody has rebuilt, you are spending into queries you can’t win. We start with the feed (titles, attributes, images, structure), then layer the bidding, the seasonal plan, and the conversion work that turns a $0.66 click into an order.
Tell us your catalog, your margins, and your goals, and we’ll show you where the budget is leaking and what a feed-first program could return.
Frequently asked
Should an ecommerce store use Performance Max or standard Shopping?
Why does everyone keep saying the product feed matters more than the campaign settings?
How much can fixing product titles really change?
What does a Google Shopping click cost, and what should I expect it to convert at?
Does seasonality really need its own plan for ecommerce Google Ads?
How does AI search affect my Google Ads and product discovery?
Every figure on this page comes from a primary platform, an independent study, or a named industry source. No competing-agency stats, no made-up numbers.
- Smarter Ecommerce, State of Performance Max 2025
- Tinuiti Q2 2025 Digital Ads Benchmark (via Karooya)
- Store Growers, product title optimization research
- Store Growers, Google Shopping ads benchmarks
- Google Ads Help, Performance Max
- Tinuiti, 2024 Cyber Five Ad Trends
- Pew Research Center, AI summaries and clicks (2025)
- BrightEdge, Post Google I/O AI Overviews data
- Adobe Analytics, generative AI retail traffic (2025)
