This listicle covers 6 audience types for Shopify stores on Meta: broad/open targeting, interest stacks, customer list lookalikes, value-based lookalikes, website visitor retargeting, and engaged social audiences.
1. Broad / Open Targeting
This would have sounded crazy in 2020, but in 2026, broad targeting (no interest or demographic targeting, just age, gender, and country) is often the best-performing audience for ecommerce on Meta. The algorithm has gotten that good at finding buyers.
How it works: you set your location, age range, and gender (if relevant), and let Meta's machine learning figure out who is most likely to buy. You are essentially saying "I trust the algorithm, here is my creative, go find buyers."
Why it works: Meta has data on 3+ billion users. It knows who buys things online, who engages with ecommerce content, and who matches the profile of your existing customers (even without a lookalike). When you constrain the audience too much, you limit the algorithm's ability to find these people.
When to use it: if you are spending $3K+/month on Meta, have strong creative, and can generate 50+ conversions per week. Broad targeting needs data volume to work. With too few conversions, the algorithm wanders.
When to avoid it: if your product is very niche or if you have fewer than 30 conversions per month. In those cases, interest targeting or lookalikes give the algorithm a helpful starting point.
2. Interest Stack Audiences
Interest targeting is not dead, but it works differently now. Instead of creating narrow audiences around one specific interest, stack 3-5 related interests into one ad set to create a broad pool (ideally 5-20 million people).
For a Shopify store selling organic skincare, an interest stack might include: Sephora, Clean Beauty, Organic food, Wellness, Yoga. These are not competitors or direct product interests. They are lifestyle signals that correlate with your buyer profile.
Tips for building interest stacks:
- Go adjacent, not direct: "Organic skincare" is obvious and competitive. "Whole Foods Market" and "meditation" are adjacent interests that signal the same buyer type with less competition.
- Keep audiences above 2 million: Smaller audiences restrict the algorithm too much.
- Test 2-3 different stacks: A "lifestyle" stack vs. a "competitor" stack vs. a "media/publication" stack. See which performs best.
- Use Advantage detailed targeting: This allows Meta to expand beyond your selected interests if it finds better prospects outside your targeting. It usually helps.
Interest stacks are a good middle ground between fully broad and overly narrow. They give Meta a signal about who you want without boxing the algorithm in.
3. Customer List Lookalikes
Lookalike audiences start with a "seed" list of your existing customers and find people who share similar characteristics. For Shopify stores, this is one of the most reliable audience types because it is based on actual buyer data.
How to set it up:
- Export your customer list from Shopify (at least 1,000 emails, ideally 3,000+)
- Upload it as a Custom Audience in Meta Ads Manager
- Create a Lookalike at 1% (most similar), 1-3% (broader but still targeted), or 3-5% (wide reach)
For most Shopify stores, 1-3% lookalikes hit the sweet spot. 1% is the most targeted but can be too small for stores with niche audiences. 5%+ lookalikes start to blur into broad targeting.
Update your customer list monthly. An outdated list trained on buyers from 6 months ago might not reflect your current customer profile, especially if your product line or pricing has changed.
One caveat: lookalikes are less powerful than they were pre-iOS because Meta has less data to match against. They still work, just not as dramatically as before. Think of them as a strong starting signal for the algorithm, not a precise targeting tool.
4. Value-Based Lookalikes
A regular lookalike finds people similar to all your customers. A value-based lookalike finds people similar to your best customers. The difference can be significant for ROAS.
To create a value-based lookalike, your customer list needs a "value" column (total spend per customer). When you upload this list and create a lookalike, Meta prioritizes matching against your highest-value buyers rather than giving equal weight to every customer.
Why this matters: your best customers (top 20% by spend) are probably very different from your worst customers. A value-based lookalike trains the algorithm to find more high spenders rather than more one-time discount buyers.
How to build the list:
- Export your Shopify customers with total spend data
- Include email, phone, and total order value columns
- Upload to Meta as a Custom Audience with "Customer value" selected
- Create a 1-3% lookalike
This audience often outperforms standard lookalikes on ROAS by 20-40%, though the audience size may be slightly smaller. It is worth testing alongside your standard lookalike to compare.
5. Website Visitor Retargeting
Website visitor audiences are the foundation of your retargeting funnel. These people already know your brand, visited your site, and showed some level of interest. They convert at 3-10x the rate of cold audiences.
Key segments to create:
- All website visitors (last 30 days): Your broadest retargeting audience. Useful as a baseline.
- Product page viewers (last 14 days): Higher intent than general visitors. They looked at specific products.
- Add-to-cart (last 7 days): Highest intent. They were ready to buy but stopped.
- Initiate checkout (last 7 days): They entered the checkout flow. Something made them leave.
For detailed retargeting setup with exclusions and sequencing, read our complete retargeting setup guide.
Important: exclude recent purchasers from all retargeting segments. Showing "come back and buy" ads to someone who just bought yesterday is annoying and wasteful. Exclude purchasers from the last 7-14 days at minimum.
6. Engaged Social Audiences
Engaged social audiences are people who interacted with your Meta content (Facebook page, Instagram profile, ads) but have not visited your website. They know your brand but have not taken the next step.
You can create Custom Audiences based on:
- People who engaged with your Instagram profile (last 30-90 days)
- People who watched 75%+ of any video ad (last 30 days)
- People who saved or shared any post (last 30 days)
- People who sent a message to your page (last 90 days)
These audiences sit between cold and warm. They are not as high-intent as website visitors, but they are more familiar with your brand than a cold audience. They respond well to ads that bridge the gap between social engagement and website visit: "See what 5,000 customers think" or a specific product showcase with a clear CTA to shop.
Engaged social audiences are also useful as lookalike seeds. A lookalike based on video viewers can perform well because video engagement is a strong signal of interest that is not affected by iOS restrictions (the engagement happened on Meta's platform, not your website). For broader audience strategy, our paid social services can help build the right mix.
Frequently Asked Questions
For prospecting, aim for 2-20 million people. Smaller than 1 million restricts the algorithm. Larger than 20 million starts to look like broad targeting. For retargeting, audience size depends on your traffic, but anything above 1,000 is usable.
Test both. Advantage+ audience (Meta picks the targeting) often outperforms manual targeting for ecommerce stores with strong creative and 50+ weekly conversions. For stores with fewer conversions, manual targeting with interest stacks gives the algorithm a helpful starting point.
Monthly. Your customer profile evolves over time, and an outdated seed list produces outdated lookalikes. Set a monthly reminder to export from Shopify and re-upload.
Find Out What's Costing You
COREPPC's free audit checks your Meta Ads account in 60 seconds. Get your performance score and see exactly where to improve.
Start Free Audit
