Most advice on PPC for a Shopify store is a list of platforms and tips. This post sets out the paid-search account we build for an online store and what running it involves, taken from our own service scope and rate card. It ends with the cases where we tell a store not to spend on ads yet.
Tracking first, because every bid learns from it
We do not build a campaign until the conversions Google Ads reports agree with the orders Shopify records. Smart bidding learns from whatever the tag tells it. If a theme snippet and an app both fire on the thank-you page, the account learns from doubled sales and bids up products that never earned it.
The comparison takes an hour and needs nothing but two exports:
- From Shopify, export orders for the last thirty days with the order date and total.
- From Google Ads, pull purchase conversions for the same days using the “by conversion time” columns, so both sides are dated the same way.
- Line the two up by day and compare the count and the revenue.
- If Google shows more, look for a duplicate tag or a purchase event that fires on page reload. If it shows fewer, look at consent handling and at browsers blocking the tag.
- Fix one cause at a time and repeat until the gap is small and you can explain what is left.
On our management retainer this is part of the written scope: consent-aware, server-side tracking set up and verified before spend scales. In our account audit it is the first test, run against real orders, with every gap listed.
The account structure we run for a store
These are the five campaigns we recommend for most stores once tracking holds. Each has its own budget and its own job, so one cannot quietly spend the money meant for another.
| Campaign | Its job | How we set it up |
|---|---|---|
| Brand search | Shoppers already typing your store’s name | Kept apart, so cheap brand sales do not make the rest of the account look better than it is. We check whether anyone else bids on the name before deciding how much it needs. |
| Shopping, split by margin group | Product searches, where most store revenue starts | Products tagged with custom labels for margin, bestsellers, new arrivals and clearance. Each group carries its own target. |
| Performance Max | Reach across Google’s other placements from the same feed | Run beside Standard Shopping on purpose and held to a Shopping benchmark you can see, because it reports less about where it spends. |
| Non-brand search | Category searches that Shopping covers poorly | A handful of collection terms per ad group, each sent to its Shopify collection page, which is where a shopper comparing options wants to land. |
| Remarketing | Visitors who viewed products or left a cart | Product ads built from the feed; Meta catalogue ads from the same feed and the same margin groups. |
When products reach Google through Shopify’s sales channel, Shopify stays the source, so an edit made only in Merchant Center can be replaced when the product next syncs. We recommend correcting titles, attributes and disapprovals in the store’s own product data. A vague title or a missing size loses the auction before any bid matters.
Why margin, not revenue, sets the targets
Return on ad spend counts revenue, so two products can hit the same target and do opposite things to your profit. The break-even point is easy to work out, and it changes how the account should be split.
Illustrative: a candle set keeps 60% of its price after product and delivery costs, and a gift box keeps 15%. Break-even ROAS is one divided by the margin: about 167% for the candles and about 667% for the gift box. At a 400% target, every candle order makes money and every gift box order loses it.
Work out the break-even figure for each product group before you set a single target. Groups whose break-even is higher than the account has ever achieved belong in remarketing or organic search, not in a prospecting campaign. Where a store shares its cost data with us, we can bid on profit rather than revenue.
- 01Verified trackingGoogle Ads conversions agree with Shopify orders
- 02Clean product dataTitles and attributes corrected in Shopify, the source
- 03Margin groupsA break-even ROAS for each product group
- 04Five campaignsEach with its own budget and its own job
What running the account involves, and how it is priced
The cadence is part of the structure. On our management retainer it is written down as follows:
- Search terms cleaned and creative tested every week, with new negatives added as they appear.
- Budget rebalanced across margin groups and channels once a month.
- Weekly reporting against store revenue, with new and returning customers split.
- A named specialist for each platform, coordinated by one strategist.
- Accounts, feed and Merchant Center kept in your name throughout.
The fee follows the work, not a percentage of your budget, and media spend is billed to you by each platform. Our management retainer runs monthly, with a minimum term of 3 months.
A store account sits in the upper half of our management range, because feed work comes on top of the campaigns. Our standalone account and tracking audit takes 1 to 2 weeks. Current ranges are on our pricing page.
When paid search is not the right answer yet
Some stores should not be buying clicks yet, and a proposal that ignores that is selling hours. These are the cases where we say so and narrow the scope:
- Tracking cannot be verified. Until reported purchases match real orders, more spend only teaches the account the wrong lesson. Fix the tag first.
- The margin cannot carry the break-even ROAS. If the break-even figure is above anything the category achieves, ads will lose money however well they are run.
- Nobody is searching for the product. Search ads capture demand; they do not create it. A new kind of product needs social or content first.
- The feed is not ready. Hundreds of disapprovals or thin titles mean the first month’s budget is better spent on the feed than on clicks.
- The budget is too small to split. Five campaigns on a small budget each starve for data. One Shopping campaign and an audit may be the honest scope.
Our PPC team has written more on the signals that mean a budget increase is the wrong move: why we sometimes recommend you spend less. The full service, including Meta and Amazon from one feed, is described on our eCommerce PPC page.
