SEO vs Google Ads: How a UK Bathroom Store on Shopify Won 64% of Its Visitors Without a Paid Click

The results at a glance

The store’s earned channels brought in 64% of users and at least 46% of revenue, measured in one GA4 window with the same totals across every report. The picture below shows, channel by channel, what each source is, how many people it brought and what they spent.

Two channels show no revenue figure because the session report on screen stopped after five rows. The £43,391 of revenue those hidden rows carry sits in neither total, so both totals are floors, not ceilings.

The first user report is the fairest place to judge acquisition, because it credits each person to the channel that first brought them to the store. On that report Organic Search is the third largest source of users, ahead of every paid row except Cross-network.

What counts as earned traffic and what counts as paid in GA4?

Earned traffic is any visit the store did not pay a click for, and paid traffic is any visit that came from an ad. In GA4 the paid side is Paid Search, Paid Shopping and Cross-network, and we count all three as paid on this page, whatever the label says.

Cross-network is the row that catches people out. The Google Analytics default channel group definitions place Google Ads traffic from the Cross-network and Google owned channels network types in this group, and name Performance Max and Demand Gen among the campaigns it covers. It carries no “Paid” label, yet it is paid. Calling it organic would flatter the story and mislead anyone who reads the screenshot properly, so it stays in the paid total, even though moving it would have made our numbers look far bigger.

Why Direct sits with earned traffic

Direct sits with earned traffic because nobody paid for the click, and because it is where GA4 files visits it cannot trace. The same Google Analytics definitions class a session as Direct when the source is “(direct)” and the medium is “(none)” or “(not set)”: typed addresses, bookmarks, links in some email apps, and links opened from apps that strip the referrer.

Several AI assistant apps behave that way. A homeowner who asks an assistant which walk-in shower suits a small bathroom, then taps the store’s name, can land as Direct rather than as a referral, which is why AI and LLM citations go largely untracked in GA4. We say that openly rather than hiding it inside a bigger organic number.

Direct is not pure organic either. Some of those 48,567 people will have met the store through an ad weeks earlier and come back by typing its name. We label the row “earned, unattributed” for that reason, and the organic claims below stand on Organic Search and Organic Shopping alone.

A note on the £2.93 million in the audience report

The audience screenshot shows £2,930,571.86 in its top row, and that figure is not revenue. GA4 adds up every audience listed, and the audiences overlap, so the same sale is counted several times. The real revenue for the window is the “All Users” row, £395,073.33, which matches the channel report exactly, down to the pence.

SEO vs Google Ads: how the two compare for an online store

SEO earns visits that cost nothing per click once a page ranks, while Google Ads buys visits that stop the day the budget stops. For this store the practical difference showed up in three places: cost per visit, how the visits behaved, and what was left when spend fell.

Why the two work better together

The two channels feed each other, which is why this store rebalanced rather than switched Ads off. Google Ads Help lists landing page experience as one of the three components of Quality Score, alongside expected clickthrough rate and ad relevance, and defines it as how relevant and useful the landing page is to people who click the ad. The same category and product pages we rebuilt for organic search are the pages the ads land on, so the SEO work also improved the pages the remaining ad budget pays to send people to.

How did the store cut its Google Ads budget by 75%?

The store cut its ad budget by moving money away from searches it now wins organically, not by switching campaigns off. The owner reports that monthly Google Ads spend fell from about £5,000 to about £1,250 over the six months, and that sales held up.

We ran the reallocation in three passes, and each pass waited for the organic rankings to hold first:

1. Overlap check. We matched the paid search terms report against Search Console and listed every product and category query where the store already ranked in the top three organically.
2. Budget pull. Paid bids on those overlapping queries were reduced first, because the store was paying for clicks it would largely have received for free.
3. Keep what only paid can do. Budget stayed on new product launches, seasonal lines and the Cross-network campaigns such as Performance Max, which still find buyers the organic pages have not yet reached.

That third pass is why Cross-network is still the largest single revenue row. A trimmed, better aimed Cross-network budget is part of the result, and we would not hide it.

Why the SEO retainer costs the owner nothing extra

The owner reports that the retainer is set at roughly 1% of the store’s annual turnover, and that the £3,750 a month released from the ad budget covers it. Put simply, the SEO is funded from money the owner was already spending on ads. The money moved from rented clicks to owned rankings.

That trade is the position we defend on every ecommerce account. Organic search is not a cost line next to paid ads. It is where the ad budget goes once the pages rank well enough to earn the clicks without paying for them, and getting them to rank is the part the method below does.

The semantic SEO method behind the numbers

The organic side was built with semantic SEO, which means organising the whole site around the store’s central entity, its shower and sauna range, so Google understands what the store sells and which page answers which search. The semantic SEO and topical authority programme at True SEO Consultants Ltd runs the same way on every site, and on this store it covered four layers: the topical map, the collection pages, the product pages and the AI citation work.

The topical map for a bathroom store

A topical map is the plan of every topic the store must cover, and which page covers it, drawn up before any page is rewritten. For this retailer the central entity is the shower and sauna range, the core section is the collection and product pages that sell, and the outer section is the guides, comparisons and planning advice that answer questions before a purchase.

We built the store’s map from three sources: Search Console queries, the product feed attributes and the questions competitors already answer. Each attribute of the range, such as size, fitting type, material, power supply and price band, was given one owner page. That stops two pages competing for the same search. Our guide to how to create a topical map shows the same method step by step with a worked example.

The map decided three things that changed the numbers:

• Which collections deserved their own page and which belonged as a filter or a heading on a parent collection.
• Which guides fed which collections, so every informational page passed its relevance to a page that sells.
• Where internal links went, from guides up to collections and from collections down to their best products.

Collection page optimisation

Collection pages, Shopify’s name for category pages, were rebuilt to answer the buyer’s question before showing the grid, because they answer the commercial searches typed by people ready to buy. Each one now opens with a short answer-first introduction naming what the collection contains and who it suits. Products come next. A buying guide sits lower down, where the shopper who is still comparing will scroll to it.

The work on each collection followed the same checklist:

• Unique introduction of two to four sentences, written from the attribute map, not from a template.
• Comparison table of the main product types in the collection, with the attributes shoppers filter by.
• Buying guide block below the grid, answering the three to five questions the collection’s searches ask.
• Controlled filters, so size, finish and price filters cannot create indexable duplicate URLs.
• Links out to the guides that explain the collection and to the sub-collections beneath it.

In the window, the seven busiest pages show bounce rates between 5.9% and 29.8%, and the busiest page drew about 16K views from about 5.2K active users with a 15.6% bounce rate.

Product page optimisation

Product pages were rewritten so that each one states facts a shopper and a search engine can both use, instead of repeating the manufacturer’s description found on dozens of other sites. Duplicate manufacturer copy gives Google no reason to rank one store’s page above every other store carrying the same text.

Each product page now carries:

• An original description built from the product’s attributes and the questions buyers ask about it.
• A specification table with the attributes in the same order as the collection filters.
• Product FAQs drawn from customer service questions and Search Console queries.
• Product structured data with price and availability, plus review data where the store collects reviews. Google’s product structured data documentation says this markup can show price, availability, review ratings, shipping and returns information in search results.
• Descriptive image names and alt text that describe what each photograph shows.

Organic Shopping shows the strongest result of the five channels in the session report. Google’s free listings, which Google Merchant Center help describes as appearing at no cost across Search, the Shopping tab, Images, Lens and YouTube, brought in 2,787 users and £20,317 of revenue at a 4.81% key event rate, the best conversion rate of the five.

What changes when the store runs on Shopify

Shopify handles part of the technical layer for you, which moves the work onto content and structure. Shopify’s help centre says every store automatically generates its sitemap.xml and robots.txt files and adds canonical tags to prevent duplicate content appearing in search results. What Shopify cannot decide is which collection owns which search, what each collection says above the grid, or how guides link to the pages that sell. That is where the hours went on this store, alongside the product feed that powers the free Shopping listings. Our Shopify SEO service for Shopify, WooCommerce and Shopwired stores runs the same checklist.

AI and LLM citation work

AI assistants cite a store when its product facts are clear, consistent and repeated across the web. So the AI work was about being easy to quote. We ran it alongside the SEO as part of our AI and LLM search optimisation service, which gets a brand named when shoppers ask an assistant which product to buy.

The work had four parts:

1. Consistent entity facts. The store’s name, range, delivery terms and returns policy read the same on the site, the Merchant Center feed and the main directories.
2. Extractable answers. Guides and collection buying blocks open each section with a one sentence answer an assistant can lift.
3. Question coverage. We covered the follow-up questions assistants generate for a product search, the pattern we explain in our piece on how query fan-out decides which pages an AI answer draws on.
4. Manual measurement. Much AI traffic lands as Direct, so True SEO Consultants Ltd tests a fixed set of buying prompts in the main assistants each month and logs whether the store is named.

What the engagement data says about content quality

The engagement data shows people reading, comparing and coming back, which is what good collection and product content should produce. Across the store, visitors spent an average of 1 minute 10 seconds engaged and triggered 1.2 million events in the window.

Three figures stand out:

• Organic Search sessions average 8.11 events, the highest of the five channels shown, which means more product views, filter clicks and basket actions per visit.
• Returning users are 11.45% of users but carry £266,209.85 of revenue, about 67% of the £395,073 total, with sessions averaging 5 minutes 4 seconds.
• The store was busy at the moment of capture. The realtime report showed 23 active users and 38 page views in 30 minutes, spread across 30 different pages.

The returning user figure matters most for a bathroom store. A homeowner who plans a shower or sauna project over weeks finds the store through a helpful guide, leaves, and comes back to buy, often by typing the name, which is one more reason Direct belongs in the earned conversation.

Should an online store switch Google Ads off completely?

No, and this case shows why. Cross-network still brought in £167,148 of revenue in the window, 42% of the total, on what the owner reports is a far smaller budget than before.

The right question is which searches you should stop paying for. Paying for clicks on searches you already win in the top three organic positions is where most of the waste sits, and that is where this store’s budget came from. Launches, seasonal lines and new audiences still justify paid spend until organic pages exist for them.

A worked sum shows the trade. A store paying 60p a click for 5,000 clicks a month on queries where it already ranks first organically spends £3,000 a month for traffic it would largely get anyway. Moving half of that budget into collection and product content that ranks for new queries turns a monthly cost into a lasting asset.