Meta Ads for eCommerce: how to structure campaigns

Short answer: The best structure for Meta Ads for eCommerce today is simple: one main Advantage+ sales campaign that gets most of the budget, one small testing campaign for new creative, and optional catalog ads when you have a large product range. Fewer campaigns and ad sets give Meta more conversion data per ad set, which helps delivery. Most of the work now goes into tracking, audience segments and creative, not into splitting audiences.

Why campaign structure changed for online stores

A few years ago, a typical store account had separate campaigns for cold interests, lookalikes, website visitors, add to carts and past buyers. Each one had its own small budget. That approach made sense when you had to tell Meta who to reach.

Two things changed. First, Meta folded automation into the normal campaign setup. In February 2025 it announced that Advantage+ would apply to sales, app and leads campaigns by default, and Advantage+ shopping campaigns became Advantage+ sales campaigns. Second, Meta rebuilt how it picks ads. Its Andromeda retrieval system was designed to handle a much larger volume of ad creative and match it to people.

The practical result: the creative now does much of the targeting. Splitting a modest budget across many ad sets usually starves each one of data. Jon Loomer, a well known Meta Ads educator, recommends consolidating budgets into fewer campaigns and ad sets, aiming for at least 50 optimised actions per week per ad set.

The Meta Ads for eCommerce structure I recommend

For most stores, I start with three campaigns at most. Here is how they fit together.

A small store with one hero product may only need the main campaign plus testing. A store with hundreds of SKUs usually benefits from catalog ads as well. What I avoid is a separate retargeting campaign by default. Advantage+ audience already reaches recent visitors and engaged users, so a standalone remarketing campaign often just competes with the main one for the same people.

How to set up the main sales campaign

Before anything else, check that purchase tracking is working through the Meta Pixel and the Conversions API, with purchase values passing correctly. If tracking is broken, structure does not matter. My guide to conversion tracking with GA4 and GTM covers the analytics side.

1. Choose the Sales objective and optimise for purchases, or for purchase value if you have enough volume and a wide price range.
2. Keep the three Advantage+ settings on: Advantage+ campaign budget, Advantage+ audience and Advantage+ placements. As Jon Loomer explains, switching any one of them off turns it into a manual sales campaign.
3. Use one ad set to start. Add a second only when there is a clear reason, such as a different country with different pricing.
4. Add audience suggestions, not restrictions. Give Meta your customer list or website visitors as signals, but avoid hard age or gender limits unless a product truly requires them.
5. Load 5 to 15 different ads. Mix formats and angles: product demo video, user generated content, static offer, carousel and a founder story.
6. Set a budget the campaign can learn on. Work backwards from your target cost per purchase so the ad set has a realistic chance of reaching steady weekly conversion volume.

I go deeper on the settings themselves in my Advantage+ sales campaigns guide.

New customers vs existing customers

This is where many store accounts lose money without noticing. Automated campaigns are good at finding people who were going to buy anyway, such as past customers and people who just left your site. That makes ROAS look great while new customer growth stays flat.

The old existing customer budget cap from Advantage+ shopping is gone, according to Jon Loomer’s breakdown of the change. What you can still do is define audience segments. In your ad account’s advertising settings, you tell Meta who your engaged audience and existing customers are, using custom audiences such as website visitors and customer lists. Then you can use the audience segments breakdown in Ads Manager to see how spend and results split between new, engaged and existing people, as described in this guide to audience segments.

My recommendation:

• Define audience segments before you launch, and update the customer list regularly.
• Check the segment breakdown every week, not just total ROAS.
• If too much spend goes to existing customers, consider a separate ad set that excludes them, and judge results on new customer revenue.
• Compare Meta’s numbers with your store’s own new customer data. Blended metrics like MER help here, which I cover in ROAS vs MER vs CAC.

Where catalog ads fit

Advantage+ catalog ads (once called dynamic product ads) pull products straight from your catalog. PPC Land describes them as ads that automatically decide which products to show each person based on how likely they are to convert. In 2025 Meta also began turning on dynamic media by default for new catalog ads created through its API, so product videos in your catalog can be used too.

Catalog ads work best when:

• You sell many products and no single product carries the store.
• Your product feed is clean: good titles, accurate prices, in stock items and sharp images.
• Your pixel events pass content IDs that match the catalog, so Meta knows which product each person viewed.

You can run catalog ads as their own campaign or add them as one of the ads inside the main sales campaign. For smaller budgets, I prefer adding them inside the main campaign so the budget is not split further.

How to test creative without breaking the main campaign

New ads added straight into a mature campaign often get little spend, because Meta favours ads with history. A separate testing campaign gives new concepts a fair chance. A simple routine:

1. Put a small fixed budget into a testing campaign with the same objective and optimisation event.
2. Launch new concepts in batches, each one a genuinely different angle, not the same video with a new headline.
3. Give each batch enough time and spend to get meaningful purchase data.
4. Move winners into the main campaign and pause clear losers.

For a full method, see my post on Facebook ad creative testing.

Common structure mistakes in store accounts

• Too many ad sets: ten ad sets at a small daily budget each rarely leave learning.
• Heavy reliance on retargeting: it looks profitable but often takes credit for sales that would have happened anyway.
• Restricting age, gender or placements without data to back it up.
• Judging ads one by one instead of looking at how the ad set performs as a whole.
• Constant edits: changing budgets and ads every day resets learning and makes results noisy.
• Ignoring the landing page: slow product pages and weak offers waste good traffic. That is where conversion rate optimisation pays off.

If you would rather have someone build and run this for you, look for Facebook Ads management that starts with tracking and audience segments, not with a long list of interest audiences.

Frequently asked questions

Need help structuring your store’s Meta Ads?

If your account has grown into a maze of campaigns, or ROAS looks good while new customers stay flat, I can help. As a Meta Ads specialist, I simplify the structure, fix tracking and set up a creative testing routine. Start with a free ads audit and I will show you what to keep, merge or switch off.

Sources

• Search Engine Land: Meta simplifies Advantage+ campaign setup, adds leads campaigns
• Engineering at Meta: Meta Andromeda
• Jon Loomer: The principles behind my Meta Ads strategy
• Jon Loomer: Advantage+ sales replaces Advantage+ shopping
• Jon Loomer: A guide to audience segments
• PPC Land: Meta enables dynamic media by default for catalog ads