How to Automate Snapchat Ads with Claude AI and MCP — AR Intelligence, Story Performance, and Real Results

Every major ad platform claims to reach young audiences. Snapchat actually does it — and the numbers are not close. Snapchat reaches 90% of 13–24 year olds and 75% of 13–34 year olds in the US. That 460 million daily active user base opens the app an average of 30 times per day. For DTC brands, lifestyle companies, and consumer products targeting Gen Z and young millennials, Snapchat is not an experimental channel — it is the primary one.

But Snapchat's ad platform behaves differently from every other paid media channel. The metrics that matter — swipe-up rate, story completion rate, AR lens engagement, share rate — are not the same KPIs you track on Meta or Google. The campaign hierarchy uses different terminology: campaigns contain ad squads (not ad sets or ad groups), and creatives live in a separate library. And Snapchat's most powerful ad format — AR Lens ads — has no equivalent anywhere else in paid media.

The challenge is that Snapchat Ads Manager surfaces these metrics across multiple views. Getting a clear picture of which ad squads are efficient, which creatives are fatiguing, and where AR is outperforming standard video requires navigating between reporting tabs, exporting CSVs, and cross-referencing manually. Most advertisers skip that work and optimize on surface metrics alone.

The Snapchat Ads MCP server changes that. It connects Claude directly to your Snapchat Ads account via the Marketing API, so instead of exporting reports and building pivot tables, you ask Claude which creatives are stalling, which ad squads need budget shifted, and whether your AR Lens is justifying its production cost — and get answers from live data in seconds.

---

What Is the Snapchat Ads MCP Server?

The Snapchat Ads MCP server implements the Model Context Protocol — the open standard that lets AI models connect to external tools and data sources with live, authenticated access.

It connects Claude to the Snapchat Marketing API, giving it direct access to your ad account data. Ask which ad squads are under-pacing — Claude queries live campaign data. Ask whether your AR Lens creative is outperforming your standard video — Claude pulls swipe-to-purchase metrics for each. Ask to pause a draining ad squad — Claude executes the change immediately (with CRUD-enabled servers).

Snapchat's campaign hierarchy has three levels: campaigns sit at the top and define the objective and budget, ad squads control targeting and bidding (equivalent to "ad groups" on Google or "ad sets" on Meta), and ads contain the creatives. The MCP server exposes all three levels plus standalone reporting and audience management endpoints.

---

Server Options Compared

1. CData — Free Read-Only GitHub Server

The CData Snapchat Ads MCP server on GitHub is a local read-only server built on CData's JDBC drivers. It wraps the Snapchat Marketing API and exposes live campaign, ad squad, and reporting data to Claude through a simple MCP interface — no SQL required. Free to use, but limited to read operations (no pausing, updating, or creating).

Best for: Developers and technical marketers who want free, local access to Snapchat Ads data for analysis and reporting without any write operations.

2. CData Connect AI — Managed, Full CRUD

CData Connect AI is CData's managed MCP platform that extends the free GitHub server with full create, read, update, and delete capability. It handles token management, server uptime, and security — you connect once and use it through Claude Desktop. Supports write operations including pausing ad squads, updating budgets, and modifying campaign status.

Best for: Teams who need full campaign management (not just read access) without running local infrastructure.

3. Adzviser — Managed, Multi-Platform

Adzviser supports Snapchat Ads alongside LinkedIn, TikTok, Pinterest, Meta, and Google — making it the best option if you're running campaigns across multiple platforms and want a single MCP connection. Authenticate via OAuth, connect your Snapchat account, and Claude gets live access to your Snap data in the same conversation as your other ad platforms.

Best for: Marketing teams running Snapchat alongside other platforms who want cross-channel analytics in a single Claude session.

4. Insightful Pipe — Analytics-Focused, Multi-Platform

Insightful Pipe connects Snapchat Ads to Claude (and ChatGPT, Gemini) alongside 30+ other platforms. It is focused on analytics and reporting — particularly useful for agencies managing multiple client accounts, with workspace-level client organization and AI persona customization per brand.

Best for: Agencies managing multiple Snapchat Ads accounts who want structured client reporting and AI-generated insights across brands.

---

Tool Surface — What Claude Can Access

Snapchat's Marketing API organizes ad management across campaigns, ad squads, ads, and creatives — plus dedicated reporting and audience endpoints. Here is what Claude can access through the MCP server:

Campaign Management

Tool	What It Does
get_campaigns	List all campaigns with objective, status, daily budget, and lifetime spend
create_campaign	Launch a new campaign with objective (awareness, consideration, conversion) and budget
update_campaign	Modify campaign status, budget cap, or name
get_campaign_stats	Pull campaign-level performance: impressions, swipes, spend, ROAS, CPM

Ad Squad Operations

Tool	What It Does
get_ad_squads	Retrieve ad squads with targeting, bid strategy, placement, and delivery status
create_ad_squad	Create an ad squad with audience targeting, bid type, and schedule
update_ad_squad	Adjust bid, budget, targeting, or delivery status
get_ad_squad_stats	Ad squad performance: swipe-up rate, CPM, CPSU, story completion rate, ROAS

Ad & Creative Management

Tool	What It Does
get_ads	List all ads with creative reference, status, and delivery data
get_creatives	Access the creative library — video, image, AR Lens, Collection, Story ad assets
update_ad	Pause, resume, or modify ad status and creative reference
get_ad_stats	Ad-level performance including video quartile views, share rate, and swipe-up rate

Performance Reporting

Tool	What It Does
get_stats	Pull performance data at any level with custom date ranges — impressions, swipes, ROAS, CPM, CPSU, video completions, story completion rate, share rate, conversions
get_product_catalog_stats	Dynamic Product Ads reporting — performance by individual product SKU within a catalog

Audience Management

Tool	What It Does
get_audiences	List custom audiences (SAM — Snap Audience Match, Pixel Custom Audience, Lookalike)
get_pixel_domains	Verify Snap Pixel installation and event firing status
get_targeting_dimensions	Browse available interest categories, age ranges, and location targeting options

---

Setup: Two Paths to Connection

Path 1 — CData Connect AI (15 Minutes, Managed, Full CRUD)

1. Go to cdata.com/ai and create an account
2. Select Snapchat Ads from the connector library and authenticate via OAuth
3. Copy your CData Connect AI MCP server URL
4. Add it to your Claude Desktop config at ~/.config/claude/claude_desktop_config.json

{
  "mcpServers": {
    "snapchat-ads": {
      "url": "https://your-workspace.cdata.com/mcp/snapchatads?auth=YOUR_TOKEN"
    }
  }
}

Path 2 — CData GitHub Server (Free, Read-Only)

Prerequisites: Java 8+ (for CData JDBC Driver). You'll need Snapchat Marketing API access.

1. Clone the repo: git clone https://github.com/CDataSoftware/snapchat-ads-mcp-server-by-cdata
2. Download the CData JDBC Driver for Snapchat Ads from cdata.com/drivers/snapchatads
3. Authenticate via the Snapchat Marketing API OAuth flow to get your access token
4. Configure the server with your Snapchat credentials and ad account ID
5. Add to your Claude Desktop config

{
  "mcpServers": {
    "snapchat-ads": {
      "command": "java",
      "args": ["-jar", "/path/to/snapchat-ads-mcp-server.jar"],
      "env": {
        "SNAPCHAT_ACCESS_TOKEN": "your_access_token_here",
        "SNAPCHAT_AD_ACCOUNT_ID": "your_ad_account_id_here"
      }
    }
  }
}

---

Workflow 1 — Morning Performance Audit with Snapchat-Specific Metrics

Here is what the actual conversation looks like:

You: "Give me a performance summary of all active Snapchat campaigns
for the last 7 days. Flag any ad squad with swipe-up rate below 1%
or CPSU above $1.50. Highlight anything spending more than $150/day
with ROAS below 2."

Claude pulls data via get_stats and get_ad_squads, then responds:

"Here's your 7-day Snapchat Ads summary across 5 active campaigns:

🔴 NEEDS ATTENTION:
- 'Summer Drop — Snap Ads' (Ad Squad: 18–24 US) — Swipe-up rate: 0.6%,
  CPSU: $1.82, Spend: $190/day. Creative is not converting the view
  into a swipe action. Story completion is healthy at 61%, suggesting
  the video holds attention but the CTA isn't driving action.
  Recommend testing a new end-card with a stronger offer.

- 'Retargeting — Site Visitors' — ROAS: 1.4, Frequency: 5.2.
  Audience is saturated. Same users seeing the ad repeatedly.
  Recommend refreshing creative or expanding to a Lookalike audience.

✅ PERFORMING WELL:
- 'TOF — Lifestyle Interest Stack' — Swipe-up rate: 2.1%, ROAS: 4.6,
  CPM: $9.20. Strong performer. Budget constrained at $200/day —
  candidate for 25% increase.

- 'Collection Ads — Product Feed' — ROAS: 5.1, 3 products driving
  80% of purchases. Recommend isolating top SKUs in a dedicated ad squad."

Workflow 2 — Story Completion Rate Analysis: Diagnosing Where You Lose the Audience

Story completion rate is one of Snapchat's most informative signals — and one of the most underused. A Snap Ad that plays to completion tells you the creative holds attention from first frame to last. A story completion rate below 25% on a 10-second ad suggests the hook isn't working. A rate below 50% on a 6-second ad suggests a fundamental creative mismatch with the audience.

Snapchat's own benchmarks put the average story completion rate at 65% for Stories content. For paid ads, the average full-screen video completion rate is approximately 42%. The gap between these numbers tells you how much room you have to improve.

Mapping completion rate against swipe-up rate reveals a two-dimensional picture of creative health: high completion + low swipe-up means the video is engaging but the call-to-action is failing. Low completion + low swipe-up means the hook itself needs a complete rethink.

You: "For all active ads in 'TOF Prospecting', show story completion
rate and swipe-up rate side by side for the last 14 days.
Segment the analysis: which ads have high completion but low swipe-up?
Which have low completion? What's the pattern?"

Claude calls get_ad_stats with video and swipe metrics:

"Story completion vs. swipe-up analysis — TOF Prospecting (14 days):

HIGH COMPLETION / LOW SWIPE-UP (CTA problem):
- Ad A (Hook: Product demo, 10s): Completion 71%, Swipe-up 0.8%
  The video is engaging — people watch it through. But the CTA
  ('Shop Now' overlay at 8 seconds) isn't generating action.
  Recommend testing a more urgent offer in the end frame:
  'Get 20% off today only' vs. generic 'Shop Now'.

- Ad C (Hook: Creator UGC, 15s): Completion 63%, Swipe-up 0.9%
  Same pattern. Long-form content holds attention but doesn't
  convert to swipes. Consider a shorter 6s cut with direct CTA.

LOW COMPLETION / LOW SWIPE-UP (Hook problem):
- Ad B (Hook: Brand logo reveal, 8s): Completion 31%, Swipe-up 0.4%
  Losing viewers in the first 3 seconds. Brand-forward opening
  is typical of low completion on Snapchat — users swipe past
  before the value proposition appears. Recommend pausing.

STRONG PERFORMER:
- Ad D (Hook: Price/offer reveal first frame, 6s): Completion 78%,
  Swipe-up 2.4%. Lead with the offer immediately — this is working."

Workflow 3 — Budget Reallocation by Ad Squad ROAS

A beauty brand was running $1,500/day across five Snapchat ad squads targeting different age and interest combinations. After three weeks, two ad squads were generating 74% of all conversions at well below target CPA — but the budget was still distributed evenly because no one had done the analysis. The other three were spending $300/day each with ROAS below 1.8.

You: "Analyze the last 21 days across all ad squads in 'Beauty — TOF'
campaign. Which are below our target CPA of $32? Which are above?
Show ROAS, spend, and conversion count for each. Recommend reallocation."

Claude analyzes via get_stats at ad squad level:

"21-day CPA performance — 'Beauty — TOF' campaign:

BELOW TARGET — scale candidates:
- F18–24, Beauty Interest: CPA $19.80, ROAS 5.2, 94 conversions → $500/day
- F25–34, Wellness Interest: CPA $24.60, ROAS 4.1, 71 conversions → $400/day

ABOVE TARGET — reduce or pause:
- M18–34, Lifestyle: CPA $58.40, ROAS 1.6, 12 conversions → Reduce to $150/day
- F35–44, Health & Fitness: CPA $71.20, ROAS 1.3, 8 conversions → Pause
  21 days, no improvement trend. Audience mismatch for this offer.
- F18–34 Broad (no interest stack): CPA $44.10, ROAS 1.9 → Reduce to $200/day

Recommended reallocation: $500 + $400 + $150 + $200 = $1,250/day.
Freeing $250/day to test a Snap Audience Match retargeting ad squad
targeting past purchasers — typically highest ROAS segment."

The brand implemented the reallocation and added the retargeting ad squad. Result: weekly conversions increased from 185 to 247 — at $250/day lower total spend.

Workflow 4 — AR Lens vs. Standard Ad Performance (Snapchat's Unique Advantage)

No other major ad platform has an equivalent to Snapchat's AR Lens ads. While Meta has some AR capabilities, Snapchat's AR is deeply native — 250 million Snapchatters engage with AR Lenses every day, and 63% of daily users use an AR filter. AR Lens ads have their own performance characteristics: they generate 2.4x higher ad awareness lift and 6.4x higher swipe-to-purchase ratios compared to standard commercial formats (Snapchat for Business data).

The challenge is that most advertisers running AR Lens ads never formally compare their performance against standard video ads in the same account. The comparison lives across different reporting views in Ads Manager. With Claude connected via MCP, it becomes a single prompt.

You: "Compare AR Lens ads vs. standard Snap Video ads across our account
for the last 30 days. Show swipe-up rate, CPSU, ROAS, and share rate
for each format. Is our AR Lens investment paying off vs. standard video?"

Claude synthesizes via get_ad_stats segmented by creative type:

"AR Lens vs. Standard Video — Account-wide (last 30 days):

                    AR Lens Ads    Standard Video
Swipe-up Rate       2.9%           1.4%
CPSU                $0.52          $0.89
ROAS                5.8x           3.1x
Share Rate          4.2%           0.8%

AR Lens ads are outperforming across every metric. The share rate gap
(4.2% vs. 0.8%) is particularly significant — Snapchat users are
sharing your AR Lens with friends at 5x the rate of standard video.
This is earned media on top of paid reach.

The CPSU delta ($0.52 vs. $0.89) means AR is 41% more efficient at
driving swipe actions — likely because users are actively engaged
with the Lens interaction before the swipe prompt appears.

Recommendation: Increase AR Lens budget allocation from current 20%
to 40% of total Snapchat spend. Prioritize Lens production for your
next product launch — the share rate makes it the only Snapchat
format with organic amplification built in."

Workflow 5 — Dynamic Product Ads and Collection Ad Performance

For e-commerce brands, Snapchat's Dynamic Product Ads (DPA) and Collection Ads are the highest-intent formats. DPA retargets users who viewed specific products with those exact products — powered by your catalog feed. Collection Ads show a video or image at the top with four product tiles below, letting users tap directly to individual product pages without leaving the ad experience.

Snapchat's Marketing API provides product-level reporting for DPA — meaning Claude can tell you not just which campaigns are performing, but which specific SKUs are driving purchases and which catalog items are consuming budget without converting.

You: "Pull the last 30 days of Dynamic Product Ads performance
across our catalog. Which product categories have ROAS above 4?
Which SKUs are spending the most with the worst conversion rate?
And for Collection Ads — which product tiles are getting the most taps?"

Claude calls get_product_catalog_stats and get_ad_stats:

"DPA + Collection Ads analysis — last 30 days:

TOP PERFORMING PRODUCT CATEGORIES (DPA):
- Sneakers (Retargeting): ROAS 7.2, avg CPA $18.40 — best performer
- Outerwear (Retargeting): ROAS 5.8, avg CPA $23.10 — strong
- Accessories (Prospecting): ROAS 4.3, avg CPA $28.60 — above target

BUDGET INEFFICIENCY (DPA):
- Basics/T-Shirts (Prospecting): ROAS 1.4, 142 clicks, 3 purchases.
  Low-intent prospecting for commoditized products. Recommend pausing
  prospecting and shifting to retargeting-only for this category.

COLLECTION ADS — tile tap analysis:
- Tile 1 (Featured item): 58% of all taps
- Tile 2 (Bestseller): 24% of all taps
- Tiles 3–4: 18% combined

Recommendation: Swap Tiles 3 and 4 for the next bestsellers in the
Sneakers category — currently the highest-ROAS segment. If Tile 1
dominates attention, feature your highest-margin SKU there, not
your most popular one."

---

Benchmarks: What the Data Shows

---

What Claude Cannot Do — Be Realistic

The right mental model: Claude is a fast, thorough analyst who can read every metric across your Snapchat account simultaneously and surface what needs action — but the creative direction, AR production, and final approval remain yours. The biggest gains come from teams who use Claude to eliminate manual reporting work and focus human time on creative strategy and audience development.

---

Combining Snapchat Ads MCP with Other Servers

Snapchat rarely runs in isolation. The most powerful setups pair it with other MCP-connected data sources Claude can query in the same conversation:

---

Getting Started Today

Frequently Asked Questions

Related Guides

• How to Automate TikTok Ads with Claude AI and MCP
• How to Automate Meta Ads with Claude AI and MCP
• How to Automate Pinterest Ads with Claude AI and MCP
• How to Automate LinkedIn Ads with Claude AI and MCP
• How to Automate X / Twitter Ads with Claude AI and MCP

---

Snapchat Ads MCP integrations are unofficial tools not affiliated with or endorsed by Snap Inc. Always review Snapchat's Marketing API terms of service for your use case.

---

Resources:

• Snapchat Marketing API Reference — Official API documentation from Snap
• CData Snapchat Ads MCP Server (GitHub) — Free, read-only, open source
• CData Connect AI: Snapchat Ads MCP — Managed, full CRUD support
• Adzviser Multi-Platform MCP — Snapchat + other ad platforms in one connection
• Insightful Pipe — Analytics-focused, agency-friendly multi-platform MCP
• Snapchat 2025 Performance & AI Updates — Snap's own platform benchmarks
• MCP Playground Online — Test MCP servers in your browser