--- title: "Snapchat Conversions API Integration" description: "Send server-side conversion events to Snapchat's Conversions API for a Snap Pixel. Agents post page views, add-to-cart, sign-up, and purchase events with customer PII automatically SHA-256 hashed before it leaves FlowRunner." url: https://flowrunner.ai/integrations/snapchat-conversions date_modified: 2026-08-04T10:04:16-07:00 --- # Snapchat Conversions API [Advertising](https://flowrunner.ai/integrations/category/marketing-social) Send server-side conversion events to Snapchat's Conversions API for a Snap Pixel. Agents post page views, add-to-cart, sign-up, and purchase events with customer PII automatically SHA-256 hashed before it leaves FlowRunner. [Verified](https://flowrunner.ai/integrations/verified "What does verified mean?") · 2 actions · API key · available [Snapchat Conversions API website](https://snap.com/) · [Platform Documentation](https://developers.snap.com/api/marketing-api/Conversions-API) · Capability data verified 2026-08-12 1. A completed order arrives from the storefront 2. The agent verifies a strong matching signal exists: email, phone, sc\_click\_id, or IP plus user agent 3. Track Purchase sends the PURCHASE event with value, currency, and the order ID as the transaction ID 4. The transaction ID doubles as the dedup key, so the same order fired by the browser pixel is not counted twice 5. The response logs to the reconciliation sheet against campaign delivery 6. Rejected events post to the marketing channel with the missing signal named 7. When refund volume against reported purchases crosses a threshold, a marketer reviews whether the event mapping is inflating the revenue signal the campaigns optimize on ## What This Integration Enables Every ad platform's bidder is a student of the conversion data you feed it, and browser pixels feed it less every year. This connector moves Snapchat conversion reporting server-side: FlowRunner agents post page views, add-to-cart, checkout, sign-up, and purchase events for a Snap Pixel directly from workflow events, so attribution reflects what your backend knows rather than what the browser managed to report. - Report purchases with value, currency, content IDs, and transaction IDs for revenue-optimized delivery - Send any standard event through the generic action: PAGE\_VIEW, VIEW\_CONTENT, ADD\_CART, START\_CHECKOUT, SIGN\_UP, and more - Recover conversions lost to ad blockers and iOS restrictions - Attribute offline and app conversions back to ad clicks via sc\_click\_id - Deduplicate against the browser pixel with a shared order ID Customer PII never leaves FlowRunner raw: email, phone, external ID, and mobile advertising ID are SHA-256 hashed with Snapchat's normalization rules before sending, and already-hashed values pass through unchanged. The human's role here is guarding the integrity of the signal itself, which is where [human-in-the-loop](https://flowrunner.ai/concepts/human-in-the-loop) review lands. ### Without FlowRunner **Conversions die in the browser**: Ad blockers and iOS restrictions eat pixel events, and Snapchat optimizes on a fraction of real purchases **Hashing PII is your problem**: Sending customer data server-side means normalizing and hashing emails and phones correctly, every time **Browser and server double-count**: Firing both paths without a shared key inflates conversions and misleads the bidder ### With FlowRunner **The server reports what actually sold**: Purchases and high-intent events post from the workflow, immune to what the browser blocked **PII is hashed before it leaves**: Email, phone, external ID, and MAID are SHA-256 hashed with Snapchat's normalization rules automatically **One order, one conversion**: A shared transaction ID deduplicates browser and server events cleanly ## Use Case Scenarios ### Every completed order becomes a purchase event When a [Shopify](https://flowrunner.ai/integrations/shopify) order completes, the agent calls Track Purchase with the order value, currency, and customer email, using the order number as the transaction ID. The browser pixel may or may not have fired; the server event arrives regardless, and the shared ID means Snapchat counts the sale once. Campaign optimization runs on real revenue instead of the ad-blocker-adjusted version. ### One event, two platforms A single conversion event in the workflow fans out: Send Conversion Event posts to Snapchat while a parallel branch feeds [Facebook Conversions API](https://flowrunner.ai/integrations/facebook-conversions-api) the same event. The hashing, timestamps, and matching signals are assembled once and delivered consistently to both bidders, which is how cross-platform attribution stays comparable instead of drifting per-channel. ### The reconciliation sheet that catches signal rot Each Track Purchase response logs to [Google Sheets](https://flowrunner.ai/integrations/google-sheets) alongside the order record. Weekly, the agent compares reported conversions against actual completed orders and posts the match rate to [Slack](https://flowrunner.ai/integrations/slack). When the rate slides, usually a missing email capture or an sc\_click\_id that stopped being stored, a person investigates before weeks of underreported conversions quietly starve the campaigns. ## Human-in-Loop Highlight Track Purchase teaches Snapchat's bidder what a customer is worth, and that lesson compounds: every event shifts how the platform spends the ad budget, and events older than 28 days cannot be corrected after the fact. The dangerous failure here is not a single bad event but a systematically wrong mapping, gross value sent where net was intended, test orders leaking into production events, refunds never subtracted. So the workflow gives a marketer standing checkpoints instead of per-event approvals: any change to the event mapping or value field ships only after review of sample events against real orders, and the weekly reconciliation flags divergence between reported revenue and booked revenue for a human to rule on. The agent fires thousands of events untouched; the definition of what counts as a conversion stays under human control, because that definition is what the money follows. Agent processes routinely Detects exception requiring judgment Clear match Continues automatically Ambiguous Routes to human via preferred channel Human decides Agent resumes with decision ## Agent Capabilities 2 actions ### Conversion Events 2 - **Send Conversion Event** Sends a single server-side conversion event for the configured Snap Pixel, covering any standard event type: PAGE\_VIEW, VIEW\_CONTENT, ADD\_CART, START\_CHECKOUT, PURCHASE, SIGN\_UP, and others. Email, phone, external ID, and mobile advertising ID are SHA-256 hashed with Snapchat's normalization rules automatically; 64-character hashes pass through unchanged. Requires at least one strong matching signal for attribution. - **Track Purchase** Sends a PURCHASE event with purchase-specific fields: value, currency, content IDs, number of items, and an order or transaction ID for deduplication. The convenience path after a completed checkout, feeding revenue into purchase-optimized ad delivery with the same automatic hashing. ## Frequently Asked Questions ### What can FlowRunner do with Snapchat Conversions API? FlowRunner agents can run Send Conversion Event and Track Purchase in Snapchat Conversions API. ### Does connecting Snapchat Conversions API to FlowRunner require OAuth? No. Snapchat Conversions API connects to FlowRunner with an API key, no OAuth flow required. ### Can Snapchat Conversions API trigger a FlowRunner workflow automatically? Snapchat Conversions API doesn't currently expose triggers in FlowRunner. It connects as an action step inside workflows started by another trigger. **Work at Snapchat Conversions API?** This integration exposes Snapchat Conversions API to AI agents on every FlowRunner plan, including through MCP, at no cost to you. [See what FlowRunner offers integration partners](https://flowrunner.ai/integrations/partners), including how to keep this page current. --- Markdown version of https://flowrunner.ai/integrations/snapchat-conversions. Site index: https://flowrunner.ai/llms.txt