Upload your email list to every ad platform in one command. Open this repo in Amp, Cursor, or VS Code and sync Customer Match, Custom Audiences, and Matched Audiences across Google, Meta, LinkedIn, TikTok, X, Reddit, and Amazon — all at once.
First-party data is the most valuable targeting signal in digital advertising. Your customer email list, when uploaded as a Custom Audience, lets ad platforms find your exact customers and build Lookalike audiences to find more people like them.
The problem? Every platform has a different audience upload format, different hashing requirements, different API endpoints, and different field mappings. Uploading a 10,000-email list to 5 platforms manually takes 2+ hours of tedious work. And audiences go stale — you need to refresh monthly.
An AI agent turns a 2-hour task into a single command. "Upload my customer list to all platforms" handles formatting, SHA-256 hashing, API uploads, and Lookalike creation across every connected platform.
Best for: Remarketing, first-party data activation, suppression lists, lookalike audience building, anyone who advertises on multiple platforms.
- 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 |
|---|---|
| Upload to All Platforms | "Sync my customer email list to Google, Meta, LinkedIn, and TikTok" |
| Lookalike Creation | "Create a 1% Lookalike on Meta from my top customers" |
| Suppression Lists | "Add my full customer list as an exclusion on all acquisition campaigns" |
| Segment-Based Audiences | "Upload VIP customers to Meta, trial users to Google" |
| Audience Refresh | "Refresh all audiences with the latest CRM data" |
| Match Rate Reports | "What's the match rate for my email list on each platform?" |
| Cross-Platform Compare | "How does my audience perform on Google vs Meta vs LinkedIn?" |
| List from CRM | "Pull my Klaviyo VIP segment and upload to all ad platforms" |
| Audience Cleanup | "Remove audiences that haven't been used in 90 days" |
You: I have 15,000 customer emails. Upload them everywhere.
Agent: I'll process your list for each platform's requirements and upload simultaneously.
Processing:
- Normalize emails (lowercase, trim whitespace)
- SHA-256 hash each email (required by all platforms for privacy)
- Format for each platform's API (field names, batch sizes, and structure differ)
Upload results:
Platform Uploaded Matched Match Rate Audience Name Google Ads 15,000 10,200 68% "Customers - March 2026" Meta Ads 15,000 11,850 79% "Customers - March 2026" 15,000 8,700 58% "Customers - March 2026" TikTok 15,000 9,450 63% "Customers - March 2026" X (Twitter) 15,000 7,350 49% "Customers - March 2026" 15,000 6,600 44% "Customers - March 2026" Match rates explained: Meta has the highest match rate (79%) because most people use their real email for Facebook. LinkedIn is at 58% because many people use a work email, not the one in your CRM. Reddit is lowest (44%) because many users use anonymous emails.
Want me to create Lookalike audiences from this list on each platform?
You: I don't want to upload one big list. I want segments for different ad strategies.
Agent: Smart. Let me build a tiered audience system:
Segment Source Size Upload To Purpose VIPs (top 10% LTV) CRM 1,500 Meta, Google Lookalike seed (highest quality) All Customers CRM 15,000 All platforms Suppression (exclude from acquisition) Cart Abandoners Shopify 2,800 Meta, Google, TikTok Retargeting (recover lost sales) Pricing Page Visitors GA4 4,200 Google, LinkedIn Retargeting (high-intent) Email Engaged Klaviyo 8,400 Meta Retargeting (warm audience) Churned Customers CRM 3,100 Meta, Google Win-back campaigns Audience hierarchy for campaigns:
- Acquisition: Target Lookalikes from VIPs, EXCLUDE all customers and cart abandoners
- Retargeting: Target cart abandoners and pricing page visitors, EXCLUDE customers
- Win-Back: Target churned customers with re-engagement offers
- Upsell: Target all customers with new product launches
Each layer has exclusions to prevent overlap — no one sees an acquisition ad AND a retargeting ad.
You: My LinkedIn match rate is only 40%. How do I improve it?
Agent: LinkedIn matches on professional email addresses, and most CRM lists contain personal emails. Here's how to improve:
Current match rate: 40% (6,000 of 15,000 matched)
Improvement strategies:
Add company names (biggest impact): LinkedIn can match by email + company name. If you have company data in your CRM, I can include it. Expected improvement: +15-20%.
Add first/last names: LinkedIn uses name + company as a secondary matching signal. Expected improvement: +5-10%.
Upload work emails separately: If you have work email addresses (from B2B lead gen), upload those as a separate field. LinkedIn matches work emails at 80%+ vs 40% for personal emails.
Combined approach:
Data Fields Estimated Match Rate Email only (current) 40% Email + Company Name 55-60% Email + Company + First/Last 60-65% Work Email + Company 75-85% You: We have company names for 80% of contacts. Add them.
Agent: Re-uploading with company names... New match rate: 62% (9,300 matched). That's 3,300 more LinkedIn members you can now target — a 55% improvement.
You: How do I keep my audiences fresh?
Agent: Audiences decay — customers churn, new buyers aren't included, and contact info changes. Here's a refresh strategy:
Monthly refresh checklist:
- ✅ Pull latest customer list from CRM/Shopify/Klaviyo
- ✅ Update VIP segment (recalculate top 10% by recent revenue)
- ✅ Update suppression list (add new customers, remove churned)
- ✅ Replace audiences on all platforms (upload new list, the platform handles diff)
- ✅ Check Lookalike audiences still based on current seed
What I'll automate:
- Monthly pull from your connected CRM (HubSpot/Shopify/Klaviyo)
- Re-upload to all 6 platforms
- Log match rates and alert if any drop significantly (could indicate data quality issues)
Why monthly matters: A 6-month-old customer list is targeting people who may have churned, changed emails, or no longer match your ICP. Fresh data = better targeting = lower CPA.
Want me to set up the monthly refresh cycle?
You: Is uploading customer emails to ad platforms legal?
Agent: It depends on your jurisdiction, but here's how the Synter MCP handles privacy:
What we do:
SHA-256 hashing: All emails are hashed before upload. The ad platform never sees raw emails. They compare hashes to their user database — if there's a match, the user is added to your audience.
No raw data stored: The Synter MCP processes emails in memory and sends hashed values. We don't store your customer list.
Platform compliance: Google Customer Match, Meta Custom Audiences, LinkedIn Matched Audiences, and all other platforms require you to have consent to use customer data for advertising.
Your responsibilities:
- ✅ Ensure your privacy policy covers using customer data for advertising
- ✅ Ensure you have consent (most Terms of Service cover this, but check yours)
- ✅ Honor opt-out requests — remove anyone who objects
- ✅ GDPR (EU): Legitimate interest OR explicit consent required
- ✅ CCPA (California): Provide opt-out mechanism
Bottom line: Customer Match / Custom Audiences are standard practice used by millions of advertisers. The hashing ensures privacy. But make sure your privacy policy and consent mechanisms cover ad targeting use.
- Upload to ALL platforms simultaneously. Your email list works on every platform. Don't leave money on the table by only uploading to one.
- Refresh audiences monthly. Stale audiences = stale targeting. New customers get added, churned customers get removed.
- VIP customers make the best Lookalike seeds. Top 10% by revenue outperforms "all customers" as a seed by 40-60% on CPA.
- Suppression saves 5-10% of budget. Excluding existing customers from acquisition campaigns is the easiest budget optimization.
- Match rates vary by platform. Meta matches 70-80%, Google 60-70%, LinkedIn 40-65% (work emails help). Don't compare platform performance without accounting for match rate differences.
- Add company names for LinkedIn. LinkedIn match rates jump 15-20% when you include company name alongside email.
- Create separate audiences for different strategies. VIPs → Lookalike seeds. All customers → Suppression. Cart abandoners → Retargeting. One-size-fits-all audience = one-size-fits-none.
Yes — this repo. It handles Customer Match (Google), Custom Audiences (Meta), Matched Audiences (LinkedIn), Custom Audiences (TikTok), Tailored Audiences (X), and more.
Emails are processed with SHA-256 hashing before upload. The ad platform compares hashes to their user database. Neither you nor the platform sees the other's raw data.
Google Ads, Meta (Facebook/Instagram), LinkedIn, TikTok, X (Twitter), Reddit, and Amazon. Additional platforms are being added.
Yes. If your CRM (HubSpot, Salesforce, Attio, Pipedrive, Klaviyo) is connected, the agent can pull segments directly and upload to ad platforms.
Monthly is the minimum. For fast-moving businesses (ecommerce, SaaS with monthly churn), bi-weekly is better.
- klaviyo-agent — Pull ecommerce segments
- hubspot-agent — Pull CRM contacts
- shopify-agent — Pull customer data
- meta-ads-agent — Meta Custom Audiences
- google-ads-agent — Google Customer Match
- linkedin-ads-agent — LinkedIn Matched Audiences
MIT — see LICENSE for details.
Built by Synter · Get API Key · Documentation