How to run Meta ads for an online store
Running Meta ads well comes down to a handful of decisions made in the right order: the numbers that define a profitable sale, a simple account, a steady supply of new ads, and an honest way to read the results. This guide takes them one at a time.
By Dan BartleyUpdated 4 minute read
Key takeaways
- Work out your breakeven ROAS before you spend: 1 divided by your contribution margin.
- Keep the account simple: one main sales campaign, broad targeting, and separate creative tests.
- New ads do most of the targeting, so plan a steady supply of them.
- Scale in steps, with new creative added before each one.
- Judge the business on store revenue and MER, and use Meta’s figures for decisions inside the account.
Set the numbers that define a profitable sale before you spend
A Meta account is judged against a target, so set it first. The two numbers that matter are your breakeven return on ad spend and the cost per purchase you can afford.
- Contribution margin. The share of each sale left after product cost, freight, fees and GST.
- Breakeven ROAS. 1 divided by that margin. A 40% margin breaks even at a ROAS of 2.5.
- Target cost per purchase. What you can spend to win a sale and still make the margin you want.
More: what breakeven ROAS do we need? and what is a good ROAS for an e-commerce brand?
Get the store and the tracking ready before the ads
Ads send people to a product page, and the page decides whether they buy. Tracking decides whether Meta learns from the sale. Both need to work before any budget goes in.
- The Meta pixel and the Conversions API both connected, through Shopify’s Meta channel or similar
- Purchases in Ads Manager roughly matching orders in Shopify
- Product pages with a full gallery, clear price, delivery and returns
- An offer you are happy to put in front of strangers
More: is our tracking broken? and the product page guide.
Keep the account structure simple
Meta learns from purchases, and every split in the account divides what it learns from. For most stores, a simple structure with most of the budget in one place outperforms a complicated one.
| Part | What it does |
|---|---|
| One main sales campaign, with Advantage+ on | Carries most of the budget and the proven ads |
| Broad, or Advantage+ audience, targeting | Lets the creative find buyers |
| A separate creative test | Gives new ads their own budget to prove themselves |
| Retargeting only where it earns its keep | Many accounts find the main campaign already covers it |
More: how should the account be structured now? and is broad targeting really better?
Plan a steady supply of new ads
With broad targeting, the ads tell Meta who they are for. An account with a few tired ads stalls, whatever the budget. New concepts, launched on a schedule, keep it finding new buyers.
- Work out how many new ads the testing budget can judge each month
- Make distinct concepts, each built around a different reason to buy
- Launch batches on a schedule, before the current winners tire
- Move proven ads into the main campaign
The full method is in the guide to how much ad creative a Meta account needs.
Scale budgets in steps, with new creative before each step
Extra budget on the same few ads mostly buys more frequency from the same people. Scale when cost per purchase has held below target for a week or more, and add new ads before each step up.
- Cost per purchase steady below target for a week or more
- Top ads not yet showing signs of fatigue
- New ads ready to launch with the higher budget
- Budget raised in steps, then held while results settle
More: why does ROAS drop every time we scale? and how do we know it is time to scale?
Judge the results on store revenue, not only on Meta’s numbers
Meta and Shopify count sales differently, so their figures never match. Use Meta’s numbers for decisions inside the account, and judge the business on what the store took.
| Number | Answers |
|---|---|
| Meta ROAS and cost per purchase | Which ads, ad sets and campaigns to back |
| MER: store revenue ÷ total ad spend | Whether the advertising as a whole pays |
| New-customer revenue | Whether the ads are growing the customer base |
More: Meta says one ROAS, Shopify another: which to trust? and which attribution setting to use.
Run Meta ads yourself, or choose someone to run them
Running the account yourself costs time; handing it over costs money and some control. Either way, keep ownership of everything and agree in writing what success means.
- The ad account, pixel and business portfolio stay in your name
- Access can be removed at any time, with no long lock-in
- Targets and the attribution setting agreed in writing
- A monthly report on cost per purchase, what was tested and what is next
- A clear answer on who makes the new ads
More: learn it yourself or hire an agency?, red flags when hiring an agency and who owns the ad account?
Questions people ask about this
How much should I spend on Meta ads to start?
Enough for steady purchases, because that is what Meta learns from. Your target cost per purchase multiplied by about 50 is roughly the weekly spend that lets an ad set settle. Below that, ads still work, less steadily.
Read the full answerHow long should ads run before I judge them?
Until each has spent enough to show a result on your target metric. A common rule is two to three times your target cost per purchase before turning off an ad that has not converted.
Read the full answerWhy did my Meta ads suddenly stop working?
Most often because the audience has seen the same ads too many times. Before blaming the ads, rule out a broken pixel, a store or checkout fault, a stock or price change and a seasonal rise in ad costs.
Read the full answerShould I hire an agency or a freelancer?
It depends on your spend and how much creative the account needs. Whoever you choose, keep the ad account in your name and agree targets in writing.
Read the full answerStill have a question? Ask us
