Google Shopping campaigns The product sells before the click.
Shopping is the only place on Google where a stranger sees your photo, your price and your shop name before they have spent a cent of your money. Which makes it the one channel where the product data is the campaign — and where most of the work happens before a bid is ever set.
Merchant Center, the feed, the campaign build and purchase tracking are one job here · We build every campaign paused, so you read the whole build before a dollar moves
Seven ways a store
loses money here.
None of these are exotic. They are what we find in almost every underperforming Shopping account we open, and not one of them shows up in a cost-per-click report.
The feed was exported, not written
Shopping matches the shopper's words against your product title. A title that reads SKU-40021 / Black loses to one that reads the brand, the product and the size — and no bid in the world fixes that.
Disapprovals nobody has opened
Merchant Center emails the warning to an address nobody checks, and a slice of the catalogue quietly stops serving. In most accounts we take over, that list has been growing for months.
One bid for four thousand products
The dozen products that pay the bills share a bid with the four thousand that do not. Shopping will not sort that out for you — somebody has to split the catalogue before the bidding means anything.
The price in the feed is not the price on the page
Google compares the two. A sale that ended, a stock level that moved or a price entered without GST sends the item to disapproved, and any click you already paid for lands on a contradiction.
Orders counted, revenue ignored
A A$19 phone case and a A$1,900 sofa each recorded as one conversion. Bidding then chases whichever is cheapest to win, which is reliably the wrong end of your catalogue.
Free listings left switched off
The same feed also places your products in the unpaid listings across Google. It is already built and it costs nothing to serve, and we still find it switched off in plenty of the accounts we open.
Performance Max quietly eating the Shopping campaign
Both campaign types read the same feed, and Performance Max outranks standard Shopping when they overlap. If nobody planned for that, the reporting stops telling you which products actually sold.
The feed is the campaign.
A Shopping ad is assembled out of your product data, inside a fixed and unforgiving shape. These are the platform limits every store is working inside — product facts, not performance claims.
150
characters per product title
The biggest single lever in the whole account. Shopping reads the shopper's query against this line, not against your ad copy.
5,000
characters per description
Room for the detail a buyer wants and the attributes Google reads. Most catalogues use a small fraction of it.
10
extra images per product
Beyond the main image. Angles, scale, the thing in use — free space most stores leave completely empty.
3
places your price must agree
The feed, the product page and the checkout. Any drift between the three and the item stops serving.
30
days before an item expires
Somebody has to refresh the product data inside that window, or the listing drops out on its own, quietly.
0
impressions a disapproved item gets
There is no partial credit. A disapproved product is out of the auction entirely until somebody fixes the reason.
So the first fortnight is spent in your catalogue, not in the ad account.
Titles get rewritten for the words buyers actually type. Missing identifiers get filled so your products can be matched against the ones they compete with. Prices, stock and shipping are made to agree across the feed, the page and the checkout, because Google checks all three.
Only then do bids matter. A perfectly tuned bid on a product Google will not serve is a decimal point being polished on a number that is already zero — which is why we do not take on a Shopping account without the feed.
Pink Flamingo · 41-day period
A$20.43
Cost per job booked
138 jobs from A$2,819.92 of ad spend
Cleanetic Perth · 15-day period
53
Jobs from a brand-new account
A$24.92 cost per job · no conversion history at launch
Cornerstone Roofing · 35-day period
68
Quote requests booked
Ready before the storm · insurance-literate
Figures are from live client accounts and are dated in each case study. These are lead-generation campaigns — we publish numbers that trace to a case study you can open, which is also why you will not find an invented return-on-ad-spend figure on this page. What carries across to a store is the method: measure the money properly, cut the waste weekly, and bid to what a sale is actually worth.
Ten things happen on a Shopping account. Here they are, in order.
This is the part most agency pages keep vague, and it is the only part that decides whether the money works. So it is written down.
Count what is actually serving
First job on any store: open Merchant Center and compare products submitted against products approved and serving. That gap is usually the cheapest revenue in the account, and it costs nothing to close.
Rewrite the titles for the way people search
Brand, product type, then the attributes people actually type — size, colour, material, model, pack quantity — front-loaded, because the start of the title carries the most weight and the end gets truncated on screen.
Fill the attributes that decide whether you compete
GTIN, brand, MPN, condition, availability, product type and Google product category. Missing identifiers do not just risk disapproval; they take you out of the comparisons where similar products get matched.
Clear the disapprovals, then keep them cleared
Every disapproval read, fixed at the source in your store rather than patched in a spreadsheet, and re-checked weekly. A feed is not a setup task — it drifts every time your catalogue changes.
Set shipping, tax and returns properly
Australian prices submitted GST-inclusive, a shipping table that matches what you actually charge, and a returns policy on file. Get these wrong and items are disapproved in exactly the regions you wanted.
Split the catalogue by margin, not by category
Your best sellers, your steady middle, your long tail and your clearance stock want different bids and different budgets. We group them by the job they do for your business, then use campaign priority and negatives to control which campaign serves which query.
Turn on everything the feed already pays for
Free listings, sale-price annotations, promotions, product ratings and local inventory where you have a shopfront. All of it runs off product data you have already built.
Track revenue, not orders
We pass the purchase value on every transaction, so a A$1,900 order does not count the same as a A$19 one. We reconcile refunds and cancellations where your platform allows it, because bidding to gross revenue overstates what you actually kept.
Bid to what a product is worth to you
We bid manually or on clicks while a new account has nothing to learn from, then set a target return on ad spend per product group once there is real purchase history. We set those targets against your margin, not against an industry number.
Read the search terms behind Shopping every week
Shopping still reports the searches that triggered it. We negate the browsing and research queries, promote the terms that buy, and keep Shopping and Search from bidding against each other on your own brand name.
The first ninety days, honestly.
- Week 0
The feed audit and the first call
Thirty to forty-five minutes on what you sell, which products carry the margin and what an order is worth. Then we open Merchant Center and your analytics, and write down every leak we can see.
- Week 1
The feed, then the campaigns
We rewrite the titles, fill the attributes, clear the disapprovals, configure shipping and tax, and prove the purchase tracking end to end. Only then do we build the campaigns. A second strategist reads all of it.
- Day one
You press go
We build every campaign paused so you can read the whole thing first — the product groups, the bids, the budgets and the negatives. You switch it on when you are ready, not when we are.
- Days 1–14
The clean
Disapprovals re-checked daily as your store syncs, search terms read and negated, and the product groups that are eating budget without selling pulled out into their own campaign where they can be controlled.
- Weeks 3–6
The catalogue sorts itself
Enough purchase data to see which products carry the account. Budget moves toward them, bidding moves from clicks to return on ad spend, and the long tail gets its own budget rather than competing with your heroes.
- From there
Rhythm
Weekly optimisation, a feed health check every week rather than every quarter, seasonal planning ahead of your peak, and reporting on a fixed cadence whether the news is good or bad.
We build every campaign paused, every time. You read the build — product groups, bids, budgets, negatives — before a dollar moves. That is a rule we hold ourselves to, not a favour.
A store account leans hardest on two of the six stages.
Compound is how our team works on any account — six stages every change passes through. Shopping puts most of its weight on measurement and on the weekly sweep.
Meter
If it can't be measured, it doesn't get bought.
On a store this is the whole ballgame. Purchase value on every order, refunds reconciled, Shopping separated from brand Search so you can see what each one earned. Until that is right, a return-on-ad-spend number is a guess with a decimal point.
Sweep
Waste is found daily, named, and priced.
Disapprovals, expiring items, price mismatches and the search terms that browse instead of buy. Each one is found, fixed and written down with a date — so “we optimised the feed” is a list you can open, not a sentence in an email.
Numbers we can show you the source of.
These four are lead-generation accounts, and every tile links to the full breakdown including the spend behind it. We would rather show you a number with a case study attached than a return-on-ad-spend figure with nothing behind it.
- End-of-lease cleaning · Melbourne138jobs booked in 41 daysA$20.43 cost per job · 21.20% conversion rate · A$2,819.92 ad spendPink Flamingo
- Carpet cleaning · Perth53jobs in the first 15 daysA$24.92 cost per job · brand-new account, no conversion history at launchCleanetic
- Roof restoration · Melbourne68quote requests in 35 daysReady before the storm · insurance-literateCornerstone Roofing
- Trades · Sydney36jobs booked in 24 daysJob-type campaign split · first quotes by day 3Sydney Fence Painting
Before you ask us anything
Here is what every store owner asks first.
Shopping, answered.
Cost, timing, contracts, who owns the account and the Merchant Center, reporting, and what happens when a campaign is not working.
The first call is free and there is no deck.
Talk to us about your accountTwo separate numbers. Your ad spend goes to Google, and what it needs to be depends on your catalogue, your margins and how competitive your products are — the smallest companies we work with invest a minimum of A$3,000 a month, and our largest clients run multiple eight figures a month across paid channels. Our management fee is separate and scales with the work; on a store the first month usually carries more of it, because the feed is rebuilt before anything is bid on. The first call is free, so you get a real number before you commit.
Shopping can produce orders in the first days, because the demand already exists and you are answering it with a product, a price and a photo. The honest caveat is the feed: if titles and attributes need rebuilding and disapprovals need clearing, that work comes first and it is worth doing properly. Most stores are in a settled rhythm around 60 to 90 days, once there is enough purchase history to bid to a return on ad spend rather than to clicks. Your margins, your competition and your catalogue size all move that.
No long-term lock-in. We would rather keep the work because it is producing than because a contract says you have to stay. What we do ask for is enough runway to be fair to the account — a Shopping campaign judged in its first ten days is being judged while the feed is still being fixed. The terms are put in front of you in plain English before anything is signed.
Merchant Center is claimed against your website, so it belongs with your business and we take an admin seat on it — your product data stays yours. The Google Ads account we own and run, because the build is proprietary methodology refined across 250+ engagements: campaign architecture, negative-keyword libraries, audience stacks and conversion tracking. You get full-visibility reporting on performance and spend at all times, and on long-term marketing agreements ownership and handover options are available to discuss.
We tell you, in the report, in plain English, and we change what we are doing. A product group that is spending without selling gets its own budget, a lower bid, or switched off. If the honest answer is that your margin cannot support paid search at the price the auction is asking, we would rather say that than keep billing you for it. What we will not do is let it run quietly and hope it comes good.
Revenue first, then the things that moved it: spend, orders, average order value, return on ad spend, and which products carried it. Each report also covers what we changed and why, and what happens next, written in the language you use about your own business. You can see live performance and spend for yourself at any time. If a number moved, you find out from us before you find it in your own accounts.
It is where your product data lives so Google can show your products — in Shopping ads, in the free listings, and in the product surfaces across Search and Images. It holds your titles, descriptions, images, prices, availability, shipping and tax settings, and it is where disapprovals are reported. We handle the setup, the website claim, the feed and the ongoing maintenance, and we fix issues in your store where the data comes from rather than patching them in a spreadsheet that goes stale.
The usual causes are a price or availability that does not match the landing page, a missing GTIN or brand, an image with promotional text on it, a restricted product category, or a website claim and returns policy that were never completed. We audit every disapproval, fix the cause at the source, and re-check weekly — because a catalogue that changes daily will drift back. The goal is a boring feed health screen, permanently.
We will not quote you a number before we have seen your margins, because it would be a guess dressed up as a promise. What we can tell you is how it gets set: we work back from what a product actually earns you after cost of goods, shipping and returns, and set the target above the point where an order stops being worth having. Then we report against that target rather than against an industry benchmark that has nothing to do with your business.
No. Shopping works with any platform — Shopify, WooCommerce, BigCommerce, Magento, Squarespace or a custom-built store. Some platforms have a direct Merchant Center connection, others need a feed file or a feed tool; either way the product data ends up in the same place. If your platform's built-in feed is producing weak titles, we override them rather than accept them, because the title is the part that decides whether you show at all.
Standard Shopping keeps the controls: product groups, bids, priorities and negative keywords, and a search terms report you can act on. Performance Max reads the same feed but hands targeting and channel distribution to Google's automation, and it outranks standard Shopping when the two overlap. Plenty of accounts run both deliberately, with brand exclusions and a clear split. What we will not do is let Performance Max absorb a working Shopping campaign and then call the result an improvement.
Often yes, and sometimes it is easier. With twenty products you can write every title properly, photograph them well and know the margin on each one, which is exactly the work that carries a Shopping account. The thing to watch is search volume: if very few people search for what you sell, Shopping alone will not create that demand, and the budget is usually better split with Search and remarketing. We will tell you which situation you are in before you spend.
We write them, you approve them. We draft the titles for the way people search — brand, product type, then the attributes buyers type — and send them to you before we submit anything, because you know your products better than we do. Where your store platform allows it we push the changes back into the store so the site and the feed stay consistent; where it does not, we override them in the feed instead.
Shopping is one of six ways to show up on Google.
Most stores we run use three or four of these together. Google Ads is the hub page that compares all six side by side, if you would rather start there.
- Performance MaxReads the same feed and outranks standard Shopping when both are live. Read this before switching it on over the top of a working account.
- Search campaignsCategory, brand and problem-shaped searches that Shopping cannot answer with a product tile alone.
- Dynamic ads for abandoned cartsDynamic ads for the exact products someone viewed, and for the carts that were filled and abandoned.
- Display networkStaying in view across the sites people use between the first visit and the purchase.
- YouTube adsShowing the product working, for the things nobody buys from a thumbnail and a price.
- The organic side of the same product catalogueFeed quality and category architecture are the same problem read twice. Fixing one usually improves the other for free.
- Counting an order as revenueHow an order becomes revenue in your reporting. Nothing on this page means anything without it.
Feed work usually comes undone at the reporting end — why the same order is counted twice, and what GST does to the value you report. Automating the routine checks also catches disapprovals before a whole feed goes quiet.
Get the feed audit
We read your Merchant Center, count how much of the catalogue is actually serving, price what the disapprovals are costing you, and show you the titles that are losing the match. Yours to keep either way.
No pitch deck. No upsell. A real conversation and a written list of leaks.
