By Your Advantage+ Shopping Campaign is showing a 5x ROAS. Your team is celebrating. But when you dig in, a question emerges:
How much of that ROAS came from people who were going to buy again anyway?
The answer, for many advertisers, is uncomfortable. Without proper caps, ASC campaigns routinely spend 40-60% of budget on existing customers. These customers already know your brand, already have purchase history, and are the easiest "conversions" the algorithm can find. They inflate your ROAS while starving new customer acquisition.
This guide shows you how to audit where your ASC budget actually goes, set proper caps, and measure whether you're growing or just re-converting the same people.
The Hidden Problem: Why ASC ROAS Is Misleading
Advantage+ Shopping Campaigns optimize for the lowest-cost conversions. Existing customers convert at 3-5x the rate of new customers. So the algorithm does exactly what you asked - it finds the cheapest conversions. They just happen to be people already in your database.
| Segment | Typical ROAS | Incremental Value | What It Means |
|---|---|---|---|
| Existing customers | 5-10x | Low (would buy anyway) | ROAS looks great but spending wasn't needed |
| Warm prospects (site visitors) | 3-5x | Medium (some lift) | Some value but overlap with organic/email |
| New customers (prospecting) | 1.5-3x | High (net new demand) | Lower ROAS but actually growing the business |
A blended ASC ROAS of 5x might actually be: 8x from existing customers (60% of spend) + 2x from new customers (40% of spend). The campaign looks great on paper but is mostly cannibalizing email and organic.
The question: Would these existing customers have bought without seeing the ad? If yes (and for repeat purchasers, the answer is often yes), that portion of your ASC spend is waste.
Why ASC Gravitates to Existing Customers
- Algorithm incentive: ASC optimizes for lowest CPA. Existing customers have lower CPA. The algorithm is doing its job - just not the job you want.
- Rich signal data: Meta has purchase history, site visit data, and engagement data for existing customers. Better data = more confident optimization.
- Higher match rates: Your customer list matches at 60-80% on Meta. Prospecting audiences have no match baseline.
- Lower risk for Meta: Serving ads to known converters lowers platform risk of wasted impressions.
- Default cap is generous: Meta's default existing customer cap (if set at all) allows substantial spend on returning buyers.
How to Audit Your ASC Spend
Follow these steps to see where your budget is actually going:
- Go to your ASC campaign → Settings → Existing customers
- If no customer list is uploaded, Meta uses its own definition (unreliable)
- Upload your full Shopify customer list for accurate identification
- Update this list monthly - stale lists miss recent buyers
- In Ads Manager, select your ASC campaign
- Click "Breakdown" → "By audience segment type"
- This shows spend, conversions, and ROAS for: Existing customers, Engaged audience, New audience
- Calculate percentage of total spend going to each segment
| Metric | How to Calculate | Red Flag |
|---|---|---|
| Existing customer % of spend | Existing spend ÷ total spend | >30% without a cap |
| New customer CPA | New customer spend ÷ new customer conversions | >3x your target CPA |
| Existing customer overlap with email | Compare existing customer purchases to Klaviyo attributed | High overlap = double-paying |
| Blended vs new ROAS gap | Blended ROAS - New customer ROAS | >2x gap means existing customers inflate heavily |
Setting the Existing Customer Cap
- ASC campaign → Settings → Existing customer budget cap
- Set as a percentage of total budget
- This limits how much Meta can spend on people in your customer list
| Business Stage | Recommended Cap | Reasoning |
|---|---|---|
| Growth-focused (acquiring new customers) | 10-15% | Maximum budget toward prospecting |
| Balanced (acquisition + retention) | 20-25% | Some retargeting value, mostly prospecting |
| Retention-focused (high LTV products) | 30-40% | Existing customers are valuable, worth re-engaging |
| Subscription business | 15-20% | Existing customers auto-renew; don't pay to re-acquire them |
The cap only works if Meta knows who your existing customers are. Without a customer list:
- Meta uses its own signals (pixel data, engagement) to identify existing customers - imprecise
- Many existing customers won't be flagged and will eat prospecting budget
- Your cap of 15% might effectively be 40% because Meta can't identify half your customers
- Upload: Shopify → Customers → Export → Upload to ASC as "Existing customer" audience
Measuring Real New Customer Acquisition
ASC labels conversions as "new" or "existing" but this is based on Meta's own identification, which can be wrong. Cross-reference with Shopify:
- Pull all orders attributed to Meta for the period (UTM tracking or Shopify source)
- Check each customer against your existing customer database
- Calculate true new customer percentage
- Compare to Meta's reported new vs existing breakdown
| Metric | Formula | Healthy Benchmark |
|---|---|---|
| New customer ratio | New customer orders ÷ total Meta orders | >60% with cap set |
| True new customer CPA | Total ASC spend × new % ÷ verified new orders | Within 1.5x your target CPA |
| New customer ROAS | New customer revenue ÷ new customer spend allocation | >1.5x (growth) or >2x (profitability) |
| Email overlap rate | Meta "new" customers who were in Klaviyo ÷ total Meta "new" | <15% (higher means bad identification) |
ASC vs Manual: The Right Budget Split
Most operators shouldn't go all-in on ASC. A hybrid approach gives you the automation benefits with manual control as a safety net:
- 60-70% of budget: Manual prospecting campaigns - Defined audiences, clear targeting, full visibility
- 20-30% of budget: ASC with cap set - Algorithm explores within guardrails
- 10% of budget: Manual retargeting - You control exactly who sees retargeting and at what frequency
- You lose visibility into what's working and why
- Budget leaks to existing customers are hard to control
- When performance drops, you can't diagnose the cause
- Manual campaigns give you a stable baseline to compare against
- You're spending >$50K/month (algorithm has enough data)
- Your existing customer cap is set and verified working
- You can cross-reference ASC conversions with Shopify new customers
- Your manual campaigns are hitting audience ceilings
Know Where Every Ad Dollar Goes
ASC's opacity is the problem. When you can't see who the algorithm is targeting, you can't know if your budget is growing the business or just retargeting your email list.
See the real breakdown.
Ask Niblin's AI agent "what percentage of my Meta spend goes to existing customers?" and get an answer computed from your Shopify customer data - not Meta's guesswork. Know your true new customer CPA. $299/mo to start.
Start Free Trial - 15 Minute Setup
Key Takeaways
- Without caps, Advantage+ spends 40-60% of budget on existing customers who'd likely buy anyway
- ASC's blended ROAS is misleading - separate new vs existing to see true acquisition efficiency
- Upload your full Shopify customer list to ASC - the cap only works if Meta can identify existing buyers
- Set existing customer cap to 10-20% for growth-focused businesses
- Cross-reference Meta's "new customer" labels with Shopify data - Meta misidentifies 15-30%
- Use a hybrid budget: 60-70% manual prospecting, 20-30% ASC with cap, 10% manual retargeting
- Track true new customer CPA, not blended CPA - this is your real acquisition cost
Frequently Asked Questions
What percentage of Advantage+ budget goes to existing customers?
Without caps, ASC typically spends 40-60% on existing customers. Meta's default settings allow substantial retargeting spend. Set the existing customer budget cap to 10-20% and upload your full customer list for accurate identification.
How do I stop Advantage+ from targeting existing customers?
You can't fully exclude existing customers from ASC - only cap spending. In ASC settings, set the existing customer budget cap to 10-20%. Upload your Shopify customer list as the existing customer audience. Without a customer list, Meta can't identify them and the cap won't work.
Is Advantage+ ROAS accurate?
ASC blended ROAS is inflated by easy existing-customer conversions. A 5x blended ROAS might be 8x from existing customers (60% of spend) and 2x from new customers. Separate the breakdown in Ads Manager to see true new customer ROAS.