Here's a fair question from a store owner: if Google Ads is an auction, why doesn't the highest bid always win?
Because the bid is only part of it. Position goes to a mix of money and fit: the price you'll pay for a click, and how squarely your ad and its page meet the shopper's question. Picture two e-bike stores in the same auction. One bids more and drops every click on its home page. The other bids less and drops the click on its folding-bike page. The second store can take the top spot.
This chapter covers how that ranking works, using Google's own description of Ad Rank and Quality Score, and what it means for an online store: build each ad group around one kind of product, and land every click on the page that sells it. The examples come from the e-bike, cookware and coffee stores in this guide. Your catalog will differ. The structure is the same.
A Store Can Outbid Every Rival and Still Lose the Top Spot. Ad Rank Counts the Page Too.
Ad Rank is Google's name for the score behind ad position. Google lists six inputs, and money buys exactly one of them:
- The bid. Whatever cap you put on a click.
- The ad's quality and the page behind it. Google weighs "the usefulness and relevance of your ad and landing page, the expectations users have about your landing page based on the clicked ad creative, ease of landing page navigation," and other quality signals.
- Ad Rank thresholds, the minimum an ad must reach to show at all.
- How competitive the auction is.
- Who's searching, and how: where they are, what device they hold, the hour, the exact words typed.
- The expected effect of your assets and ad formats, like sitelinks or a phone number.
Google doesn't publish a formula, and you'll find old blog posts that claim one. Ignore them. Google does tell advertisers that sharper ads usually win cheaper clicks. Quality is the part of Ad Rank a store can improve without raising a single bid.
"Folding Electric Bike" Deserves an Ad Group of Its Own. A Store-Wide Ad Fits No Search Well.
Of all the quality factors, ad relevance is the easiest to move, and it moves by how you group keywords. Google asks whether your ad fits what the shopper meant. An ad that has to cover everything in your store matches no particular search well.
So we build tight ad groups: five or six closely related searches that all want the same kind of product, one ad written for them, and one page to land on. Aventon's search data, from Keyword Planner, sorts cleanly this way:
- Folding: "folding electric bike" (90,500 a month), "folding fat tire electric bike" (9,900), "foldable electric bike for adults" (3,600), "lightweight folding electric bike" (880)
- Fat tire: "fat tire electric bike" (40,500), "step through fat tire ebike" (390)
- Commuter: "commuter ebike" (74,000), "best commuter ebike" (1,600)
The folding group gets an ad that says folding in the headline and lands on folding bikes. The fat tire group gets its own. A shopper who typed "lightweight folding electric bike" sees an ad that repeats what they asked for, clicks, and lands on a page full of exactly that.
One caution when you build these lists: Keyword Planner often returns several spellings of one search with the same volume, like "commuter ebike" and "electric commuter bike," both at 74,000. Google is reporting one pool of searches, not two. Keep the spellings in the same ad group and don't add their volumes together.
"Fat Tire Electric Bike" Gets Five Times the Searches of the Aventon Level 3. Its Category Page Should Catch the Click.
Store owners think in products. Shoppers search in categories. That mismatch shows up the first time you put a product list next to Keyword Planner.
In Keyword Planner, "aventon level 3," one of the brand's model names, draws about 8,100 searches a month. "Fat tire electric bike" draws 40,500. The plain word "ebike," which Keyword Planner groups with its close variants, shows about 1.5 million, roughly 185 times the model. In Aventon's data, the category searches dwarf the model search.
That decides where clicks should land:
- Category searches ("fat tire electric bike," "carbon steel pan," "colombian coffee") land on the category page that shows every matching product.
- Model searches ("aventon level 3") land on that product's page.
- The home page catches neither well.
Our example stores already have the pages. Made In has a carbon steel collection. Volcanica sorts its coffee by region, with pages for Colombian, Costa Rican and Ethiopian beans. Lifeboost organizes by roast. Each of those pages is a ready landing spot for its own ad group.
That makes the account structure almost write itself. A coffee store that sells by origin builds an ad group per origin: one for Colombian, one for Ethiopian, one for Costa Rican, each with the five or six ways shoppers ask for it and an ad that names the origin in its first headline. A store that sells by roast does the same with dark, medium and light. The category pages you already maintain are the blueprint; the ad groups just follow them.
No page for a category your ads need? Make the page first, then the ad group. Twenty minutes on a collection page with an honest paragraph and the right products saves weeks of paying for clicks that land in the wrong aisle.
Google Grades the Page Your Ad Promised, Not Your Home Page.
Landing page experience is Google's read on whether the page delivered for the person who clicked. The yardstick is that one shopper and that one ad. A home page can be beautiful and still score poorly for a shopper who asked for a carbon steel pan and got a banner for bakeware.
A store's home page answers every search a little and none of them well. It has to introduce the brand, push this week's promotion and point to every department. A category page has one job. When the ad promised folding bikes, a page of folding bikes keeps the promise.
Speed is part of the experience too, and large product photos are a common cause of slow store pages. A category page that loads forty full-size images before showing a single price makes the shopper wait for the very thing they came to see. Run your category pages through Google's PageSpeed Insights and check the mobile score, since Google asks for mobile-friendly landing pages. A page that takes too long to show its products loses people before relevance can matter. But fix relevance first: a fast page that shows the wrong products is still the wrong page.
Shopping Ads Have No Keywords, So No Keyword Quality Score. The Product Title Does That Job.
Everything above is about Search ads. Shopping ads work differently, and it's worth knowing why before Chapter 6 puts them in your first month.
Google reports Quality Score for keywords in Search campaigns. Shopping campaigns don't have keywords at all: Google says Shopping ads "use your existing Merchant Center product data (not keywords) to decide how and where to show your ads." That means the title, the description, the category, the brand and the product identifiers. So the work that tight ad groups do for Search, your product feed does for Shopping. A product titled "12 in. Carbon Steel Frying Pan, Pre-Seasoned" can match searches that a product titled "The Chef Pan" can't.
Chapter 8 treats product titles as ad copy, and Chapter 14 covers the feed itself.
Quality Score Is a Diagnosis, Not a Grade. The Weakest of Its Three Parts Tells You What to Fix.
Every keyword carries a score from 1 to 10. Google is blunt that the auction never reads it; the score exists to show you where an ad group is weak. Behind the number sit three readings, each marked above average, average or below average against other advertisers:
- Expected click-through rate: the odds a shopper picks your ad when it appears.
- Ad relevance: whether the ad matches what the shopper meant by the search.
- Landing page experience: whether the page delivers on the ad.
Read it by keyword, not as an account average. An account-wide Quality Score blends your best and worst ad groups into one number that describes neither, so it hides the one group dragging costs up. A "Below average" on ad relevance usually means the keyword sits in an ad group that's too broad; split it out. A "Below average" on landing page experience usually means the click lands on the wrong page; point it at the category. Expected click-through rate is judged with ad position and assets taken out of the picture, so raising your bid won't fix it. A sharper ad written for a tighter group will. Change one thing at a time. Google compares your keyword with advertisers on the same searches over the last 90 days, so a newly split ad group needs fresh clicks before its readings mean anything.
Next: Seeding Keyword Planner From Your Own Site Misses the Brands You Sell.
Tight groups need good lists. Chapter 5 builds them with Keyword Planner, and shows why a store that resells other brands has to seed it with those brand names, not just its own website.
Earlier chapters cover what a click can cost before it loses money and tracking the purchase. Every chapter is on the guide's home page.
When we build a store account through Google Ads management, the ad groups follow the store's category pages, not the other way around. Doing it yourself? List your top five category searches and check that each one has a page that shows only those products.
