The Hidden Cost of Wrong Attribution: A $50K Case Study

The Brand That Was Losing $50K Per Month Without Knowing It

This is the story of a DTC skincare brand spending $180K per month on paid media across Meta, Google, and TikTok. Their agency reported a blended ROAS of 3.2x. The numbers looked healthy. The CEO was happy. The media buying team was optimizing against platform-reported metrics and scaling campaigns that appeared profitable.

They were losing approximately $50K per month in misallocated ad spend.

Not because their ads were bad. Not because their product didn't sell. Because their attribution model was telling them the wrong story about where their revenue actually came from.

Here's exactly what went wrong, what the real numbers looked like, and what changed when they fixed it.

The Setup: What the Platform Data Showed

The brand's monthly budget allocation and platform-reported performance:

| Channel | Monthly Spend | Platform-Reported Revenue | Platform ROAS | |---------|--------------|--------------------------|---------------| | Meta Prospecting | $45,000 | $67,500 | 1.5x | | Meta Retargeting | $35,000 | $175,000 | 5.0x | | Google Branded Search | $30,000 | $240,000 | 8.0x | | Google Shopping | $40,000 | $120,000 | 3.0x | | TikTok | $30,000 | $60,000 | 2.0x | | Total | $180,000 | $662,500 | 3.7x |

The conclusions seemed obvious:

• Google Branded Search is the star at 8.0x ROAS -- increase budget
• Meta Retargeting is efficient at 5.0x -- maintain or increase
• Meta Prospecting is marginal at 1.5x -- consider cutting
• TikTok is underperforming at 2.0x -- reduce or pause

The agency followed this logic. Over two months, they shifted $15K from Meta Prospecting to Google Branded Search and Meta Retargeting.

What Happened Next

Within 4 weeks of the budget reallocation:

• Google Branded Search ROAS dropped from 8.0x to 5.5x despite more spend
• Meta Retargeting ROAS dropped from 5.0x to 3.2x
• Overall conversion volume declined by 18%
• Blended ROAS fell from 3.2x to 2.4x

The agency's reaction was to optimize harder -- new creatives, tighter audiences, more aggressive bid strategies. Nothing worked. The problem wasn't execution. The problem was the data that drove the budget decision.

The Root Cause: Three Attribution Failures

An independent audit revealed three compounding attribution errors.

Failure 1: Google Branded Search Was Cannibalizing Organic

Google Branded Search reported $240K in revenue at 8.0x ROAS. But the brand had strong organic rankings for its own brand terms. A quick analysis showed:

• Pausing branded search for one week resulted in only a 12% drop in brand-term traffic, not the 100% drop you'd expect if those ads were incremental
• 88% of branded search clicks would have come through organic results anyway
• The true incremental revenue from branded search was approximately $29K, not $240K -- an actual ROAS of roughly 1.0x

The brand was paying $30K/month to capture traffic that would have arrived for free.

Failure 2: Platform Double-Counting Inflated Total Revenue

Adding up platform-reported revenue: $662,500. The brand's actual revenue from Shopify: $430,000.

The platforms collectively over-reported by 54%. This is typical when running across multiple channels -- each platform claims credit for overlapping conversions. But the agency was using the $662K number for blended ROAS calculations, making overall performance look 54% better than reality.

Actual blended ROAS: $430K / $180K = 2.4x, not the reported 3.7x.

Failure 3: Meta Prospecting Was the Real Growth Driver

The most damaging error was the undervaluation of Meta Prospecting.

Platform-reported: 1.5x ROAS. This made it look like the worst-performing channel.

But Meta Prospecting was filling the top of the funnel. Those new users later:

• Searched the brand name on Google (credited to Google Branded Search)
• Were retargeted by Meta and Google (credited to retargeting)
• Returned directly and purchased (credited to direct traffic)

An independent attribution analysis using first-party data and multi-touch modeling showed Meta Prospecting's actual contribution was closer to 2.8x ROAS when downstream conversions were properly attributed to the channel that initiated the journey.

Cutting Meta Prospecting didn't just hurt that channel -- it dried up the pipeline feeding Branded Search and Retargeting, which is exactly why both those channels declined after the budget shift.

The Fix: What They Changed

Step 1: Independent Attribution

The brand implemented server-side tracking with multi-touch attribution. This gave them:

• A single conversion count (no double-counting across platforms)
• Position-based credit distribution (40% first touch, 40% last touch, 20% middle)
• Cross-channel visibility showing how touchpoints from different platforms connected

Step 2: Incremental ROAS Targets

Instead of using platform-reported ROAS, they set targets based on incremental ROAS -- the additional revenue generated by each channel that wouldn't have happened without it.

Google Branded Search shifted from an 8.0x target (platform-reported) to a 1.0x incremental target, with spend capped at the level needed to prevent competitor brand bidding.

Step 3: Budget Reallocation Based on Real Data

New allocation based on independent attribution:

| Channel | Previous Spend | New Spend | Change | |---------|---------------|-----------|--------| | Meta Prospecting | $30,000 (post-cut) | $60,000 | +$30K | | Meta Retargeting | $40,000 | $25,000 | -$15K | | Google Branded Search | $40,000 | $15,000 | -$25K | | Google Shopping | $40,000 | $45,000 | +$5K | | TikTok | $30,000 | $35,000 | +$5K | | Total | $180,000 | $180,000 | -- |

Same total budget. Fundamentally different allocation.

The Results

After 60 days of the new allocation:

• Total revenue increased from $430K to $510K per month (+19%)
• Actual blended ROAS improved from 2.4x to 2.8x
• New customer acquisition increased by 34%
• Cost per new customer decreased by 22%

The $50K "waste" wasn't spending too much overall -- it was spending too much on bottom-funnel channels that captured existing demand and too little on top-funnel channels that created new demand.

The Patterns to Watch For

This case study isn't unique. The same patterns appear across brands spending $50K-$500K/month on paid media:

Pattern 1: Branded search appears to be your best channel. If branded search shows 5x+ ROAS, test pausing it for a week. If traffic only drops 10-20%, you're mostly paying for clicks you'd get organically.

Pattern 2: Retargeting has the highest ROAS. Retargeting converts people who were already considering buying. High ROAS doesn't mean high incrementality. Reduce retargeting spend gradually and measure whether total conversions actually decline proportionally.

Pattern 3: Prospecting looks unprofitable. Prospecting campaigns create new demand that flows through other channels. If you cut prospecting and see retargeting and branded search decline 2-4 weeks later, prospecting was the engine driving those channels.

Pattern 4: Platform totals exceed actual revenue. If the sum of all platform-reported conversions exceeds your actual conversions by more than 30%, you need independent attribution to understand true channel performance.

Frequently Asked Questions

How do I convince a client that their "best performing" channel is actually wasting money?

Start with the data gap. Show them platform-reported total conversions vs actual conversions from their ecommerce platform. When they see platforms collectively claiming 50% more conversions than actually occurred, the credibility of platform-specific reporting drops. Then run a simple incrementality test: pause branded search for one week and measure the actual impact on revenue. The gap between expected loss (100% of branded search revenue) and actual loss (usually 10-20%) makes the case more effectively than any presentation.

What's a reasonable budget to spend on branded search if most of it is cannibalistic?

Only spend enough to defend against competitors bidding on your brand terms. Check if competitors are showing ads for your brand name. If they are, maintain enough branded search budget to hold the top position. If they're not, you can likely reduce branded search to a minimal presence or pause it entirely. Most brands find the right branded search budget is 30-50% of what they were spending when they believed the platform-reported ROAS numbers.

How long does it take to see the impact of budget reallocation based on better attribution?

Expect 30-60 days before the full impact is visible. Top-of-funnel changes (increasing prospecting spend) take 2-4 weeks to flow through the funnel as new users progress from awareness to consideration to purchase. Bottom-of-funnel changes (reducing retargeting or branded search) show impact faster, typically within 1-2 weeks. Run the new allocation for at least 60 days before evaluating results, and compare against the same period in the previous year to control for seasonality.

---

Go Funnel uses server-side tracking and multi-touch attribution to show you which ads actually drive revenue. Book a call to see your real numbers.