If you are looking for a Google Ads agency for ecommerce, you probably want someone to take the account off your hands and make it pay. That is part of what we do: SSO manages Google Ads campaigns for Australian physical-product stores as part of our Growth Program. What comes first is working out whether the account can pay at all, and what it has to return to cover your costs.
Where Google Ads fits in an ecommerce business
Google mostly captures demand that already exists. Someone types what they want, and your ad appears while they are looking. That is different from Meta, which puts a product in front of people who were not searching for it. Our guide to Google Ads vs Meta Ads for ecommerce compares the two in detail.
The practical consequence is that Google's ceiling is set by how many people search for what you sell, and how much you can afford to pay for each of them. Four situations come up often.
-
People already search for what you sell.
Category terms, product types and the problems your product solves. Search and Shopping ads can put you in front of those shoppers at the point of comparison.
-
People search for your brand name.
Brand campaigns can protect those searches from competitors, but much of that revenue may have arrived anyway. It needs to be judged separately from campaigns aimed at new customers.
-
Your product is new or unfamiliar.
Search volume may be thin. Google may play a supporting role while another channel creates the demand.
-
Your average order is low or your margins are thin.
Every click is paid for out of the gross profit on the order. In a competitive auction, a low-value order can lose money on every sale however well the campaign is built.
That last situation is where many accounts go wrong, and it is not specific to Google. In January 2026, Apothecary by Justin was paying about $400 to buy a customer worth about $55 in gross profit. No campaign structure closes a gap like that. The offer had to change first, and only then did the advertising have something it could pay for.
Search, Shopping and feed-related responsibilities
Google offers several campaign types, and they depend on different inputs. In plain terms, these are the three an ecommerce store usually meets.
-
Search ads.
Text ads that appear against search terms. Control comes from which terms you bid on, which you exclude, what the ad says and where the click lands.
-
Shopping ads.
Product listings with an image, title and price, shown when someone searches for a product. They are built from product data supplied through Google Merchant Center, not from keywords you choose.
-
Performance Max.
A campaign type in which Google's automation decides where ads appear across its inventory, including Shopping placements, using your product data and creative. It relies on accurate conversion data and sensible limits, because the reporting shows less about what is driving results.
The product feed is the foundation of Shopping
A product feed is the list of your products with the details Google needs: title, description, price, availability, images, shipping, and identifiers such as brand and GTIN where they exist. Most store platforms can produce one through a built-in sales channel or an app, and it is submitted through Merchant Center.
The feed decides which searches your products can appear for, and whether they appear at all. A price on the feed that differs from the product page, a product listed as in stock when it is not, or a vague title written for the store rather than the shopper can each mean disapproved products or ads that show for the wrong searches.
| Area | What gets checked | Usually sits with |
|---|---|---|
| Product information | Titles, descriptions and images are accurate, specific and match the product page | The store owner, who knows the product and controls the store |
| Price, stock and shipping | The feed matches what the shopper sees at the moment of purchase | The store platform and whoever manages the store |
| Merchant Center account | The account is verified, and disapprovals and policy warnings are cleared | The store owner as account holder, with whoever maintains the feed |
| Conversion tracking | Each purchase is recorded once, with the correct value | Shared, and confirmed before campaigns spend money |
| Campaigns, bids and search terms | Budget goes to products and terms that can pay for themselves | Whoever manages the Google Ads campaigns |
Who holds each responsibility in a specific engagement is written into the scope before work starts.
When we manage Google Ads campaigns for a client, the proposal sets out the work we take on and the decisions, approvals and information we need from you, before you commit. Feed maintenance is one item that has to be named there explicitly, because it sits between the store and the ad account and is easy to leave with nobody.
If Shopping ads are underperforming, check the feed before the bids. A campaign cannot show a product the feed describes badly, or leaves out.
Measuring results beyond a platform headline
Google Ads reports the conversions it can connect to its own ads. That figure is useful for running the account. It is not the same thing as profit, and it is not the same as what Google added to your store.
-
Channels overlap.
Google, Meta and email can each claim the same order. Add the platform figures together and they often come to more than the store actually sold.
-
Brand search flatters the average.
Searches for your own name tend to show very high reported returns. Blending them with category campaigns hides whether new customers are being bought at a profit.
-
A return has to clear what your costs require.
Whether a figure is good depends on your margins and costs, not an industry number. Our guide on what a good ROAS is for your business shows how to work out the return you need.
-
The average hides the next dollar.
The last $1,000 added to a budget usually returns less than the first. Deciding when to increase your ad budget is a question about that next increment, not the account average.
So alongside the platform reports we look at store-level revenue against total media spend, the break-even return for your cost base, what is left after media, and how many customers are new rather than returning. MER and ROAS answer different questions, and both are needed.
| Measure | Figure | What it tells you |
|---|---|---|
| Google-reported revenue | $30,000 on $6,000 spend (5.0x) | What Google claims, including brand search and orders other channels also claim |
| Brand campaigns | $18,000 on $1,000 spend | Much of this demand existed before the click |
| Non-brand campaigns | $12,000 on $5,000 spend (2.4x) | Closer to the cost of reaching new buyers |
| Break-even return for this store | 2.5x | The non-brand campaigns sit just below break-even on this basis |
Hypothetical figures for illustration.
At 5.0x, the account looks healthy. The question that matters is whether the campaigns aimed at new customers pay, and on these figures they do not quite. That changes the next move: look at the products being pushed, the offer or the landing pages before adding budget.
Scope, evidence and the Sprint entry
Google Ads management is one part of the work we deliver inside the Growth Program, alongside the offer, the website, email and any other channel the plan needs. Every engagement starts with a four-week Validation Sprint, including one where Google Ads is the main piece of work. We do not take on work we have not measured.
- 1
Your numbers first
Week one works out what the business has to earn: cost of goods, shipping, returns and what you pay yourself. That sets the return any channel, Google included, has to clear.
- 2
Existing data or new data
If you already have trading and ad history, we audit what is there and run one targeted test. If there is no sales data yet, campaigns go live to buy some. That test budget is additional to the Sprint fee, and agreed and capped before anything spends.
- 3
Test against what the data shows
Week three follows what moves and stops what does not.
- 4
A written answer
In week four we spend ninety minutes with you on every number, where it came from, and the recommendation in writing.
The Sprint is $4000 + GST, paid upfront, and runs four weeks from the day we have your access. There is no refund if you do not like the finding. If we go on to an initial twelve-month programme together, the full $4000 is credited across invoices 11 and 12, $2,000 each. It does not reduce the first ten invoices.
A Sprint ends in one of three ways: an offer to do the work, a fix list for later, or no offer and a clear reason. Roughly six in ten completed Sprints are not followed by an offer from us (September 2026 record, not a promised future rate). A positive finding does not guarantee an offer.
If we do go ahead, the route follows what the Sprint found, not your turnover. Launch is for economics that need building or repair: twelve months, 30% of agreed eligible product gross profit against a $1,500 plus GST monthly minimum, and you pay the greater of the two, never both. Accelerate is for economics that already work and need capacity: a twelve-month initial agreement, invoiced monthly in advance and scoped in writing. Our overview of the work SSO delivers covers the other channels, and our client results show what the work has produced for other stores, with the costs attached.
What to have ready for a Sprint where Google matters
- Access to your Google Ads account, if one exists
- Access to Merchant Center, if you run or plan to run Shopping
- Access to your store platform and analytics
- Cost of goods, shipping, returns and payment fees by product line
- What you pay yourself, and what you need the business to produce