How I Saved a $5M Ecommerce Company from a Facebook Disaster

(and how you can too!)

In August 2024, what started as a typical month for a thriving $5M ecommerce company quickly turned into a nightmare scenario. The data tells the story: a dramatic increase in sessions, plummeting conversion rates, and declining sales threatened to derail years of steady growth. Here's how we identified the root cause and implemented a solution that not only stopped the bleeding but set the stage for a remarkable recovery.

The Crisis Unfolds

The numbers were alarming:

  • Sessions spiked from about 7,000 to 14,000+ overnight

  • Conversion rates plunged from a robust 3% to below 1%

  • Daily sales decreased from $15K to starting to fall below $5K

Upon detecting unusual metrics, I immediately dove into both Google Analytics and Shopify metrics to identify the source. The data pointed to Meta - Facebook had inexplicably begun delivering a massive increase in impressions, resulting in elevated click-through rates. While this surge in traffic could have been advantageous, the critical issue was that these visitors weren't converting. 

“Upon detecting unusual metrics, I immediately dove into both Google Analytics and Shopify metrics to identify the source.”

My initial response was persistent outreach to Meta, making five separate attempts at contact. Despite our account's significant advertising investment - nearly $1.5M in the past year alone - we received no meaningful response. A thorough audit of ad placements revealed nothing unusual; the spike was uniformly distributed across all placement types. Given that we hadn't implemented any website changes or new applications, the evidence strongly suggested a Meta-side issue. To test this hypothesis, I created a new ad account while maintaining the same pixel, only to encounter identical problems. Frustrating. 

Tips, tactics and value delivered to your inbox:

The nuclear option loomed: creating an entirely new Facebook page and pixel from scratch - a daunting prospect I was reluctant to pursue. Before taking such drastic measures, I decided to investigate one final angle. Since our campaign's inception, we'd observed persistent attribution discrepancies: Facebook Ad Manager consistently reported metrics at double their actual values - twice the ROAS, double the conversion value, and cost per purchase at half the true amount. Following my e-commerce principle of "if it's not broken, don't fix it," I'd previously left this attribution anomaly untouched given the account's stellar performance. Now, it demanded investigation. This led me to schedule a consultation with the founder of wetracked.io (I owe a huge debt of gratitude to Asad). 

Our investigation revealed a critical issue: our Shopify website attribution system was misreporting data, causing Meta to make misguided advertising decisions based on incorrect performance metrics. Here’s what we did to fix it: 

The Three-Pronged Solution

1. Facebook Data Sharing Reset

The first step was to address the root cause. We discovered that the Shopify sales channel app for Facebook and Instagram was sharing conflicting data. By turning off data sharing and reconfiguring it correctly, we ensured that Facebook received accurate conversion data. (Facebook and Instagram sales Channel App→ Settings→Share Data-Off). 

2. Google Shopping Feed Optimization

The secondary source of data discrepancy was in Google Ads, specifically through the Symprosys Google Shopping feed. We switched this app from the primary to secondary source of conversion data to ensure it aligned with the actual customer journey and product performance.

3. Implementation of Advanced Tracking

To prevent future discrepancies and provide a single source of truth, we integrated wetracked.io software with both Shopify and all advertising platforms. This gave us:

  • Real-time accurate attribution

  • Cross-platform data verification

  • Improved decision-making capabilities

The Recovery

The results were almost immediate:

  1. Immediately, conversion rates began trending upward, reaching 2% and continuing to climb:

  1. Sales stabilized and showed consistent growth:

  1. Sessions were trending downward in the account because we drastically pulled back ad spend. But, you can see at the end of September (when we implemented changes), they started to fall even more, while the quality of traffic improved significantly:

Key Learnings

  1. Multiple data feed apps can overcount data: Regular audits of attribution systems are crucial

  2. Unified tracking matters: Having a single source of truth prevents costly mistakes

  3. Quick action saves dollars: The faster you can identify and fix data issues, the less revenue you'll lose

Moving Forward

With corrected attribution and improved tracking in place, the company is now poised for sustainable growth. The crisis served as a catalyst for implementing better systems and processes, ultimately making the business stronger and more resilient.

Remember: In ecommerce, you’re at the mercy of ad platforms. When in doubt, go back through the entire customer journey to look for errors and even potential issues.

Subscribe Below to Download a Seller’s Discretionary Summary Spreadsheet

Subscribe to keep reading

This content is free, but you must be subscribed to Digital Acquisition Club to continue reading.

Already a subscriber?Sign In.Not now