Bridge email marketing and paid advertising for maximum ROAS. Open this repo in Amp, Cursor, or VS Code and manage Klaviyo lists, segments, and revenue data with AI — sync your top customers to ad platforms for lookalikes and suppress email subscribers from acquisition campaigns.
Klaviyo is the email/SMS marketing platform of choice for 143,000+ ecommerce brands. It tracks every email open, click, purchase, and browse event — giving it the richest understanding of customer behavior in the ecommerce stack.
This data is transformative for advertising. Your Klaviyo segments reveal exactly who your best customers are: top 10% by revenue, repeat purchasers, VIP tier, recently churned. Each segment becomes a high-quality ad audience.
The smartest ecommerce brands use Klaviyo and ad platforms in a feedback loop: ads acquire new customers → Klaviyo nurtures them → purchase data feeds back into ad audiences → better targeting → lower CAC. An AI agent automates this entire loop.
Best for: Ecommerce and D2C brands, Shopify stores, any business using Klaviyo for email/SMS that also runs paid ads on Meta, Google, or TikTok.
- Get a free API key at syntermedia.ai/developer
- Set the key:
export SYNTER_API_KEY=syn_your_key_here - Open this repo in your editor
- Start chatting — MCP tools are pre-configured in
.mcp.json
Copy claude_desktop_config.json to your Claude config directory and replace the API key:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
| Capability | Example |
|---|---|
| Revenue Segments | "Find my top 10% customers by lifetime revenue" |
| Lookalike Seeds | "Sync VIP customers to Meta for a 1% Lookalike audience" |
| Customer Suppression | "Exclude all Klaviyo subscribers from my Meta acquisition campaigns" |
| Purchase Behavior | "Which customers have purchased 3+ times in the last 90 days?" |
| Churn Segments | "Find customers who purchased 6+ months ago but not since" |
| Flow Performance | "How is my welcome series performing? Open rates, click rates, revenue?" |
| Campaign Analytics | "Which email campaign drove the most revenue this month?" |
| Segment Building | "Create a segment of customers who opened emails but haven't purchased" |
| List Management | "How many subscribers are on my main newsletter list?" |
| Revenue Attribution | "What percentage of revenue comes from email vs SMS vs ads?" |
You: I want to find more customers exactly like my best buyers. What data from Klaviyo should I use?
Agent: Klaviyo has the richest customer data for building Lookalike audiences. Here's my tiered approach:
Tier 1 — Highest quality (your VIP seed):
- Segment: Top 10% by lifetime revenue + purchased 3+ times + last purchase within 90 days
- Size: 1,240 customers
- Upload to Meta → 1% Lookalike (~2.3M people)
- These are your absolute best customers. The Lookalike will find people with similar purchase patterns.
Tier 2 — Repeat buyers (scale audience):
- Segment: Purchased 2+ times + average order value above $50
- Size: 3,890 customers
- Upload to Meta → 2% Lookalike (~4.6M people)
- Slightly broader than Tier 1, good for scaling after Tier 1 is exhausted
Tier 3 — All buyers (maximum scale):
- Segment: At least 1 purchase
- Size: 12,400 customers
- Upload to Google Ads → Customer Match + Similar Audiences
- Broadest audience, used when you need volume
Expected impact: Tier 1 Lookalikes typically deliver 40-60% lower CPA than interest targeting. Your VIP customers have behavioral patterns (browsing, engagement, purchase frequency) that Meta's algorithm can match.
You: Build all three tiers and sync to Meta.
Agent: Done. Three Custom Audiences created on Meta with corresponding Lookalikes. I've also set up your 12,400 buyer list as an exclusion audience on your acquisition campaigns — no point paying to reach people Klaviyo already reaches for free.
You: Am I wasting money showing ads to people who are already on my email list?
Agent: Let me check the overlap between your Klaviyo subscribers and Meta ad audiences...
Overlap analysis:
- Klaviyo total subscribers: 48,200
- Meta acquisition campaign reach (last 30 days): 890,000
- Estimated overlap: 6,200 people (0.7% of reach)
The math on waste:
- Average CPM: $14.00
- Impressions to overlapping subscribers: ~31,000
- Wasted spend: ~$434/month
But here's the nuance: Not all overlap is waste.
- Subscribers who opened email in last 7 days → already engaged, don't need ads → SUPPRESS
- Subscribers who haven't opened email in 90+ days → "email dead" → KEEP showing ads
- Subscribers who abandoned cart → RETARGET (ads + email double-down works)
My recommendation:
- Suppress "active email subscribers" (opened email in last 30 days) from acquisition campaigns → saves $350/month
- KEEP targeting "email dead" subscribers — ads are the only way to reach them now
- Create a dedicated "cart abandoner" retargeting campaign that complements your Klaviyo abandonment flow
You: What percentage of our revenue comes from email vs ads?
Agent: Pulling Klaviyo revenue attribution and comparing to your ad platform data...
Last 30 Days Revenue Attribution:
Channel Revenue % of Total Cost ROAS Klaviyo Email (Flows) $48,200 32% $450 (platform cost) 107x Klaviyo Email (Campaigns) $22,800 15% $450 (platform cost) 50x Klaviyo SMS $8,400 6% $1,200 (SMS cost) 7x Meta Ads $34,600 23% $12,000 2.9x Google Ads $24,200 16% $8,500 2.8x Organic / Direct $12,800 8% $0 ∞ Key insight: Klaviyo drives 53% of revenue at a tiny fraction of the cost. But here's the catch — many of those email customers were originally acquired through ads. The real picture:
Customer journey: Meta Ad → First Purchase → Klaviyo Welcome Flow → Second Purchase → Klaviyo Campaign → Third Purchase
The ad platforms don't get credit for purchases 2 and 3, but they made them possible. Your true acquisition cost should factor in the lifetime value that email unlocks.
Recommendation: Don't cut ad budget because "email drives more revenue." Instead, optimize ads for first-purchase acquisition and let Klaviyo handle retention. The combo is what drives your 53% email revenue.
You: I want to win back customers who stopped buying.
Agent: Let me identify churned customers in Klaviyo...
Churn segments:
Segment Criteria Count Avg Previous LTV Recently Lapsed Purchased 3-6 months ago, not since 2,840 $120 Lapsed Purchased 6-12 months ago, not since 1,920 $95 Deeply Lapsed Purchased 12+ months ago, not since 3,100 $72 Multi-channel win-back strategy:
Email (Klaviyo flow — you should already have this):
- 3-email sequence: "We miss you" → "Here's what's new" → "10% off, last chance"
Ads (complement the email flow):
- Recently Lapsed (2,840): Highest recovery potential. Upload to Meta → show new product launches and bestsellers. Budget: $15/day.
- Lapsed (1,920): Medium potential. Upload to Meta → show "We miss you — 15% off" offer with new products they haven't seen. Budget: $10/day.
- Deeply Lapsed (3,100): Low potential but large audience. Upload to Google Display → low-cost brand awareness. Budget: $5/day.
Why ads + email is better than email alone: Your email open rate for lapsed customers is probably 10-15%. Ads reach the other 85-90% who won't open your email. The double-exposure increases win-back rates by 30-40%.
You: I just launched SMS marketing on Klaviyo. How should it work with my ads?
Agent: SMS and ads should complement each other, not compete. Here's the coordination framework:
Rule 1: Never run the same offer on SMS and ads simultaneously to the same audience.
- If you SMS "Flash Sale: 20% off today only" to your full list, suppress that list from ad platforms for 24 hours. Otherwise you're paying to tell people something they already know.
Rule 2: Use SMS for urgency, ads for awareness.
- SMS: "Your cart is expiring in 1 hour — complete your purchase"
- Ads: "New arrivals you'll love" (discovery and browsing)
Rule 3: SMS subscribers make the best ad audiences.
- People who opted into SMS are your most engaged customers
- Upload SMS subscribers to Meta → 1% Lookalike → find more high-engagement customers
- SMS subscribers who also purchase → best seed audience you'll ever have
Coordination calendar:
Day SMS Ads Monday — Product discovery ads Tuesday — Product discovery ads Wednesday New Arrival Alert Suppress SMS list from ads Thursday — Retarget non-openers with same products Friday Flash Sale 🎉 Suppress SMS list from ads Weekend — Retarget with sale items
- Top 10% by revenue = best Lookalike seed. This segment outperforms "all customers" as a Lookalike seed by 40-60% on CPA. Quality > quantity for seed audiences.
- Suppress active email subscribers from acquisition. If someone opens your emails regularly, don't waste ad budget reaching them. They're already engaged.
- "Email dead" subscribers need ads. Subscribers who haven't opened in 90+ days won't see your emails. Ads are the only channel that can re-engage them.
- Cart abandoner double-down. Run Klaviyo's abandonment flow AND retargeting ads simultaneously. The multi-channel approach recovers 30-40% more carts than email alone.
- Revenue attribution is misleading. Email "drives" revenue, but ads acquired the customer. Track first-purchase source separately from repeat-purchase source.
- Refresh audiences monthly. Customer purchase behavior changes. Last month's VIPs might be this month's churners.
- SMS subscribers = highest-intent Lookalike seed. People who give you their phone number are deeply engaged. Use this segment as your premium Lookalike source.
Yes — this repo. It pre-configures the Synter MCP server for Klaviyo management. Works with Amp, Cursor, VS Code, and Claude Desktop.
Yes. The agent exports Klaviyo segment data, hashes emails, and uploads as Meta Custom Audiences for Lookalike creation.
It complements it. Klaviyo's native Meta integration syncs lists, but the AI agent adds intelligence — choosing the right segments, building suppression logic, and coordinating timing.
The agent focuses on reading segments, lists, and performance data for ad targeting. Flow creation is best done in Klaviyo's visual builder.
By using Klaviyo's rich purchase data to build better ad audiences (Lookalikes from VIPs), suppress waste (don't advertise to active subscribers), and coordinate channels (email + ads together).
- shopify-agent — Shopify product & customer data
- meta-ads-agent — Meta Custom Audiences & Lookalikes
- google-ads-agent — Google Customer Match
- audience-sync-agent — Multi-platform audience sync
- cross-platform-ads-agent — Multi-channel reporting
- tiktok-ads-agent — TikTok Custom Audiences
MIT — see LICENSE for details.
Built by Synter · Get API Key · Documentation