Google Ads for E-Commerce: Structuring Shopping Campaigns That Convert
Search campaigns get most of the attention in Google Ads discussions, but for online stores it is Shopping that usually carries the budget — and the revenue. A well-structured Shopping campaign puts your product, price and photo directly in front of someone who has already decided to buy something like it. The catch is that Shopping campaigns are built differently from Search: the product feed does most of the targeting work, and the account structure exists mainly to control how budget flows to different parts of that feed. Get the feed and structure wrong, and no amount of bid tweaking will fix it.
This guide walks through how to structure Shopping campaigns for an e-commerce catalogue of any size — from a 40-SKU boutique store to a several-thousand-product marketplace — and where the money is usually being wasted. It draws on patterns that show up across most e-commerce accounts regardless of vertical: the same feed gaps, the same flat budget splits, and the same neglected search terms reports tend to appear whether the store sells furniture, electronics or clothing.
Why Shopping campaigns behave differently from Search
In a standard Search campaign, you choose keywords and Google matches them to queries. In Shopping, there are no keywords to bid on. Instead, Google matches queries against the data in your Merchant Center product feed — title, description, product type, GTIN, brand, attributes — and decides which products are relevant enough to show. Your campaign structure then determines which ad group (and which bid) a given product falls into once it has already been deemed eligible.
That distinction matters because it flips the usual order of optimisation. In Search, you start with keyword research. In Shopping, you start with the feed, because a vague or thin feed will cap performance regardless of how carefully the campaigns are organised afterward. Only once the feed is solid does campaign structure become the lever worth pulling.
The feed is the targeting
Product titles are the single biggest lever most stores leave untouched. A title like “Blue Cotton T-Shirt — Men’s, Size M” gives Google far more to match against than “T-Shirt 4521”. Include the attributes a shopper would actually search: material, colour, size, use case, brand. Custom labels are the other underused field — they let you tag products by margin, season, or bestseller status without touching the core taxonomy, which becomes essential once you start splitting campaigns by priority.
Structuring campaigns around priority and margin, not just category
The most common mistake is organising Shopping campaigns purely by product category, the way a store’s navigation is organised. Category structure feels intuitive but ignores the fact that not all products deserve the same budget treatment. A better approach layers two things: performance tier and margin, using custom labels to separate them.
A three-tier structure that scales
A pattern that works for most catalogues over a few hundred SKUs:
- Tier 1 — Bestsellers and high-margin items. A dedicated campaign with a higher daily budget and target ROAS, so these products are never budget-constrained.
- Tier 2 — The core catalogue. Everything else, grouped by product type, with a standard bidding strategy and a shared budget pool.
- Tier 3 — Clearance, low-margin or discontinued stock. A capped-budget campaign with a lower target ROAS, so these items get some visibility without eating spend that should go to Tier 1.
Google’s Shopping auction gives priority (high, medium, low) as a campaign-level setting specifically to manage overlap between campaigns targeting the same products — use it deliberately: Tier 1 as high priority, Tier 2 as medium, Tier 3 as low, so if a bestseller also qualifies under a broader campaign, the higher-priority one wins the auction.
Performance Max versus standard Shopping
Performance Max campaigns now handle a growing share of Shopping inventory automatically, pulling in Display, YouTube and Search alongside product listings. They perform well once there is enough conversion volume for the automation to learn from — generally 30 or more conversions a month per campaign is a reasonable threshold. Below that volume, a standard Shopping campaign with manual tier structure usually outperforms Performance Max simply because there is more control while the account is still building a signal.
| Factor | Standard Shopping | Performance Max |
|---|---|---|
| Control over placements | Shopping surfaces only | Search, Display, YouTube, Gmail, Maps |
| Bidding | Manual CPC, Target ROAS, Maximise Clicks | Automated only (Target ROAS / Conversions) |
| Negative keywords | Full control at campaign/ad group level | Account-level exclusions only |
| Visibility into search terms | Full search term report | Limited insights, category-level only |
| Best suited for | New accounts, tight margin control, low volume | Established accounts with 30+ monthly conversions |
| Setup effort | Higher — manual tier and priority structure | Lower — mostly feed and creative assets |
Budget allocation that matches real margin
Splitting budget evenly across campaigns is the easiest mistake to make and the easiest to fix. Pull margin data into your custom labels and let the tier structure above allocate spend accordingly — Tier 1 should typically receive 50-60% of total Shopping budget even though it may represent a smaller share of the catalogue by SKU count, because it is where the return is highest per euro spent.
Seasonal and promotional campaigns
For sales events, create a time-boxed campaign rather than editing your always-on structure. A separate campaign with its own budget, priority set to high, and a clear start/end date keeps promotional spend contained and makes post-event analysis far easier than trying to untangle a permanent campaign that had a temporary bid change applied during the sale window.
Negative keywords still matter in Shopping
Even without keyword targeting, Shopping campaigns generate a search terms report, and it is usually full of waste: browsers looking for free versions, unrelated brand names, or product types you don’t stock. Review it weekly for the first month of a new campaign structure, then monthly once patterns stabilise. This is one of the fastest ways to reduce wasted spend without touching bids at all.
For stores also running Search and Facebook simultaneously, keeping budget allocation consistent with actual channel performance rather than habit is worth revisiting regularly — our comparison of Google Ads and Facebook Ads budgets covers how to think about that split when a business runs both.
Reporting: what to check weekly versus monthly
Weekly: search terms report, disapproved products in Merchant Center, and budget-constrained campaigns (a flag Google shows directly in the campaign list). Monthly: ROAS by tier, product-level conversion rate, and whether any Tier 1 products have drifted into Tier 3 territory due to falling margin or stock issues. If your overall Google Ads budget is being reviewed at the same cadence, align both reviews so Shopping isn’t assessed in isolation from Search.
None of this replaces a properly optimised product page — a well-structured campaign will send more qualified clicks to a page, but if that page’s on-page SEO and conversion basics aren’t in place, those clicks convert at a lower rate regardless of how tight the campaign structure is.
Common mistakes that quietly drain Shopping budget
Most underperforming Shopping accounts aren’t broken in an obvious way — they’re leaking budget through a handful of small, compounding issues. These are the ones worth checking first when a campaign’s ROAS has been flat or declining.
Letting Google’s automatic bid adjustments run unchecked
Device and audience bid adjustments applied months ago and never revisited are a common source of waste. A -20% mobile adjustment made when mobile converted poorly two years ago may no longer reflect reality, especially if the site’s mobile checkout has since improved. Review bid adjustments at the same cadence as the rest of the account rather than treating them as set-and-forget.
Ignoring Merchant Center diagnostics until products disappear
Products can silently lose approval for missing GTINs, price mismatches between the feed and the live page, or policy flags — and Google doesn’t always surface this loudly. A weekly glance at the Merchant Center diagnostics tab catches this before it shows up as a mysterious drop in impressions three weeks later.
Treating the feed as a one-time setup task
Feeds decay. Prices change, stock runs out, seasonal products should rotate out of visibility, and new attributes become available as Google expands what it accepts. Stores that assign feed maintenance to whoever set it up initially, with no recurring review, tend to see slow performance decline that’s hard to trace back to a single cause because it happens gradually.
Working with a specialist versus managing Shopping in-house
Smaller stores often start by managing Shopping themselves, which is entirely reasonable while the catalogue and budget are modest — the tier structure in this guide can be built and maintained by a single person with a few hours a month. The point where outside help starts to pay for itself is usually when the catalogue crosses a few hundred SKUs, when the account is running Shopping alongside Search and Facebook Ads and the interactions between them need coordinating, or when internal time is better spent elsewhere in the business.
If you’re weighing that decision, it’s worth comparing the ongoing cost of a managed Google Ads arrangement against the actual hours a tiered, well-maintained Shopping structure takes to run properly — for many stores it’s less about whether they could manage it, and more about whether managing it well is the best use of the time it takes each month.
The Bottom Line
Shopping campaigns reward structure more than clever bidding. Start with a feed that gives Google enough signal to match well, split the catalogue into priority tiers based on margin rather than navigation categories, and use campaign priority settings deliberately when tiers overlap. Standard Shopping still beats Performance Max for accounts without enough conversion volume to feed the automation, so don’t switch just because it’s the newer option. Review search terms weekly at first, keep promotional campaigns separate from always-on structure, and revisit budget splits against actual ROAS by tier rather than leaving them static. Stores that treat Shopping structure as a living system — adjusted as margins and stock change — consistently get more revenue from the same daily budget than stores that set it up once and leave it alone.
Frequently Asked Questions
How many products do I need before Shopping campaigns make sense?
There’s no strict minimum — even a 20-product store can run Shopping profitably — but the tiered structure described above starts paying off once you have at least 50-100 SKUs, since below that a single well-optimised campaign is usually simpler to manage.
Should I use manual CPC or Target ROAS bidding?
Manual CPC gives more control while an account is new and still building conversion history. Once a campaign has accumulated 15-20 conversions, testing Target ROAS on that specific campaign usually improves efficiency, because Google’s bidding algorithm has enough data to make reasonable per-auction decisions.
Why are some of my products not showing at all?
Check Merchant Center diagnostics first — disapproved items, missing GTINs, or policy violations are the most common causes. If products are approved but still not serving, low bids relative to competition or a thin product title are the next things to check.
Is Performance Max going to replace standard Shopping entirely?
Google has been pushing advertisers toward Performance Max, but standard Shopping campaigns remain available and, for lower-volume accounts, often still outperform it due to tighter control over budget and exclusions.
How often should I update product titles and descriptions?
Review titles whenever a product’s performance is noticeably below similar items in the same tier — a title rewrite is often the fastest fix. Beyond that, a quarterly pass across the catalogue to add missing attributes is usually enough.
What’s a reasonable budget split between Tier 1 and the rest of the catalogue?
As a starting point, 50-60% of total Shopping budget to Tier 1 bestsellers is reasonable, then adjust based on actual ROAS after the first month of data rather than sticking to a fixed ratio indefinitely.
Do I need a separate campaign for every product category?
No — that’s the structure-by-navigation trap this guide argues against. Group by performance tier first, and only split by category within a tier if the category has meaningfully different margins or seasonality.