Google Shopping Campaign Template
Restructuring a live Shopping account costs you the learning period twice, so the cheapest place to change your mind about campaign structure is a spreadsheet. This one holds the six structures worth considering, what each one costs you, and a build sheet you complete before anything is created in the account.
FORMAT
XLSX
CONTAINS
4 tabs
LAST UPDATED
September 2026
LICENCE
Free, unrestricted
Every structure decision is a decision about what you will be able to see later.
- Why this one
The structure decision is made once and paid for repeatedly
Most Shopping accounts are structured around a theory nobody present can still explain.
Shopping is the one channel where the ad is your catalogue data and the campaign is a budget instruction.
Campaign structure decides what you can see and what you can control. A single campaign covering the whole catalogue will spend the same acquiring a customer for a product at 60 per cent margin as one at 8 per cent, and no amount of bid management inside that campaign can fix it, because the campaign is the unit of budget.
The opposite failure is splitting on everything, which divides conversion data so thinly that automated bidding never leaves the learning period. One rule resolves both: split on anything you would budget differently, and nothing else.
- Six structures, each with what it costs you stated plainly
- A build sheet you complete before opening Google Ads
- A naming convention agreed before the first campaign is typed
- The five negative lists worth building as shared lists
- Five checks that say whether a campaign is ready to build
It covers both Performance Max and Standard Shopping, because most accounts run both and the interesting question is which one you expect to win the auction.
- What is inside
Four tabs
Small on purpose. A build sheet nobody finishes is worse than no build sheet.
Read me
The five-step order for filling it in, and five counters that read the Build tab and tell you what is not ready yet.
- Campaigns planned, total daily budget, approximate monthly
- Count of campaigns with no target set
- Count of campaigns with no negative list applied
Build
Eleven columns per campaign, pre-filled with a worked eight-campaign structure covering margin tiers, manual control and a second market.
- Name, type, priority, daily budget
- Ad group or asset group, and the product group it subdivides
- Bid strategy, target, negatives applied
- What this campaign is for, in a sentence, and its status
Structures
Six structures with how each one splits, when to use it, and what it costs you. The last column is the one most comparisons leave out.
- Single campaign, margin tier, category, bestsellers
- Priority ladder for Standard Shopping
- By market, and when that is worth the thinner data
- How to choose between Performance Max and Standard Shopping
Naming
A four-segment naming convention with worked examples, and the five negative lists to build as shared lists before launch.
- Type, market, intent, detail
- Brand exclusion, junk terms, competitors, out of scope, wrong market
- Why renaming later breaks more than it fixes
Nothing in the file is locked or protected. If a structure does not suit your catalogue, change it.
- The six structures
What each one costs you
Every structure buys control and pays for it in data density. The last column is the part most comparisons leave out.
Structure
How it splits
Use it when
What it costs you
Single campaign
One Performance Max campaign, all products
You are starting out, or the catalogue has similar margin throughout and fewer than about two hundred products
No control. You cannot spend differently on products that earn differently, and you cannot see which half is working.
By margin tier
Three or four campaigns split on a custom label carrying the margin band
Margin varies meaningfully across the catalogue, which on most stores it does
You have to maintain the custom label. If the margin data is wrong, the entire structure is wrong.
By category
One campaign per category or department
Categories have genuinely different customers, seasonality or buying cycles
Thinner conversion data per campaign. Small categories may never leave the learning period.
Bestsellers separated
One campaign for the products that carry revenue, one for the rest
A small number of products produce most of the sales, which is usual
Needs reviewing quarterly. This year bestsellers are not next year bestsellers.
Priority ladder
Three Standard Shopping campaigns at High, Medium and Low priority, steered by negatives
You want explicit control over which query buys which product, and somebody to run it
Labour, and a person who understands how priority and negatives interact. It works; it is not free.
By market
One campaign per country or currency
Delivery cost, margin or competitive intensity differs by market
More campaigns, thinner data. Only split markets you would genuinely budget separately.
The rule underneath all six: split on anything you would budget differently, and on nothing else.
- How to use it
From blank sheet to built account
Half a day for the planning, which is considerably less than a fortnight of relearning.
01
Work out your margin bands before anything else
Most of the structures on the Structures tab split on margin, and none of them work if the margin data is guessed. Landed cost, not invoice cost, and returns taken out. The break-even ROAS calculator turns each band into the target you will actually enter.
02
Pick one structure and write down why
The Structures tab states what each one costs you. Choose against your catalogue rather than against what worked on somebody else. Put the reason in the sheet, because in six months nobody will remember it.
03
Agree the naming convention before you type a campaign name
Four segments: type, market, intent, detail. Renaming later keeps the history in Google Ads and breaks every saved report and dashboard that referenced the old name.
04
Fill the Build tab and check the counters
Eleven columns per campaign. A campaign with no target and no negative list applied is not ready, and the Read me tab counts both for you.
05
Create the shared negative lists first
Five lists, built at account level so there is one place to edit them. Brand exclusion is the one that matters most, because without it you cannot separate acquisition from defence afterwards.
06
Build it, then leave it alone long enough to learn
Decide the review window before launch, not after the first bad week. Editing targets during the learning period restarts it, and an account edited weekly never produces data worth reading.
If you are restructuring rather than building new, change the structure once. Doing it in stages costs you the learning period at every stage.
- Getting it wrong
Six structural mistakes
All six are cheap to avoid on a spreadsheet and expensive to fix in a live account.
One campaign for the whole catalogue
The campaign is the unit of budget, so a single campaign spends the same on a 60 per cent margin product and an 8 per cent one. No bid adjustment inside it can fix that.
Splitting on everything
The opposite failure. Twenty campaigns across a catalogue that converts forty times a month means no campaign ever leaves the learning period.
Leaving brand inside acquisition
Brand converts at rates acquisition cannot approach. Mixed together, the average describes neither, and the account looks healthier than it is.
Naming campaigns as you go
Three months later nobody can filter the account. Renaming keeps the history in Google Ads and breaks every report built on the old names.
Setting targets before knowing break-even
A target ROAS chosen because it sounds ambitious is a decision to buy revenue at an unknown margin. Break-even ROAS is one divided by gross margin, after returns.
Restructuring in stages
Each stage resets learning. If the structure is wrong, change it once, accept a fortnight of noise, and then read the data.
If more than two of these apply, plan the whole structure on the Build tab before changing anything in the account.
DOWNLOAD
Google Shopping Campaign Template — XLSX
Six structure options with the trade-off stated for each, a build sheet pre-filled with a worked margin-tier structure, a four-segment naming convention, five negative lists to create first, and five checks that tell you whether a campaign is ready to build.
FORMAT
XLSX spreadsheet
TABS
Read me, Build, Structures, Naming
STRUCTURES
6, with trade-offs
LICENCE
Free, commercial use
No email address, no sign-up and no watermark. Eight worked example rows sit at the top of the Build tab and should be deleted before you start.
- Questions
Shopping campaign structure FAQs
What people ask before rebuilding an account.
What is the best Google Shopping campaign structure?
There is no single best one, and the honest answer is a rule rather than a structure: split on anything you would budget differently, and on nothing else. For most stores that means margin tiers, because margin is the thing that actually varies and the thing that decides what you can afford to pay for a customer.
Should I use Performance Max or Standard Shopping?
Performance Max for most of the catalogue, with brand excluded so you can measure it. Standard Shopping where you want explicit control over which query buys which product, or where you are judging a campaign on units rather than return, such as clearance. Running both is normal.
How many campaigns should I have?
As few as your budget decisions require. A useful test: if you would never move money between two campaigns, they should probably be one. Splitting divides the conversion data automated bidding needs, and thin campaigns never leave the learning period.
Do I still need the priority ladder?
Only if you want manual control and have somebody to run it. It is a Standard Shopping technique using High, Medium and Low priority with negatives to steer traffic, and it still works. It is labour, not magic, and most accounts do better putting that labour into the feed.
How do I split by margin if I do not have margin data?
Get the margin data first. Landed cost rather than invoice cost, with returns taken out. Until then any structure you build is guessing, and so is every target in it.
Why does the naming convention matter so much?
Because it is the only filter you get. Campaign names are what you segment reports by, and renaming later keeps the history inside Google Ads while breaking every saved report, dashboard and filter built on the old names.
How long before I judge a new structure?
Two to four weeks on most eCommerce accounts, and decide the window before you launch rather than after the first bad week. Changing targets during the learning period restarts it.
Does this replace feed work?
No, and if you only have time for one, do the feed. In Shopping the feed is the ad; campaign structure is a budget instruction. The feed optimisation guide covers the other half.
- Keep going
Related reading
GUIDE
The other half of Shopping, and the half worth doing first. Titles, attributes and structure.
CHECKLIST
Forty-two checks for auditing an account that already exists, starting with whether the numbers are true.
Planned the structure and want it checked?
The free audit covers the account and the feed with the specialists who manage eCommerce paid media daily, plus five other disciplines, ranked into one constraint.