TLDR;
- Your top-of-funnel ads are working, but you are losing a large portion of potential customers in the middle of the funnel, and generic retargeting is not fixing it.
- The data shows that a broad, automated approach like Google Demand Gen for retargeting can perform poorly, while a segmented sequence on Meta can be over twice as efficient as prospecting campaigns.
- The fix is to stop lumping all warm traffic together. By separating retargeting audiences by intent (such as newsletter subscribers versus page engagers), you can reduce middle-funnel acquisition costs dramatically.
- Below, interactive tools based on this account's data help diagnose funnel leaks and model the financial impact of fixing them.
I see this frequently. A growth leader comes to me after cracking top-of-funnel acquisition. Prospecting ads pull in traffic, clicks are inexpensive, and brand visibility is high. However, revenue does not scale at the same pace. A gap exists between the initial click and the final conversion, and capital is being wasted on broad retargeting campaigns that underperform. The common assumption is that retargeting is simple—showing an ad to anyone who visited the site. That exact assumption often drives up costs unnecessarily.
The core question is how to eliminate steep drop-offs in the middle of the funnel, and whether replacing an underperforming retargeting campaign—such as a Google Demand Gen campaign on YouTube—with a segmented sequence of custom audiences on Facebook and Instagram provides the solution. This requires a fundamental rethink of how we re-engage potential customers who have already shown interest.
To examine this thoroughly, raw performance data from a complete campaign run for SitterTree is analyzed. SitterTree operates a childcare services marketplace app connecting parents with babysitters, targeting parents as new users—a notoriously challenging and expensive audience to convert. This examination walks through dissecting ad account data to identify what works, what fails, and how much money is lost when the middle of the funnel is mismanaged.
The Anatomy of the Data: A Single Account Under the Microscope
All figures, calculations, and findings in this analysis originate from a single advertiser's accounts for SitterTree. The dataset covers performance across Google Ads and Meta Ads from July 2024 through early 2025. In total, the campaigns generated 5,779 signups from a combined spend of just over $22,000, representing a real-world dataset from a B2C services company.
This is not an industry-wide study or a meta-analysis of hundreds of accounts. The numbers presented are not generic benchmarks to copy and paste. Rather, this is a detailed examination of one specific case. The primary value lies in the methodology—the process of diagnosis, analysis, and strategic correction. The underlying principles apply to almost any business dealing with a leaky funnel, though specific figures remain unique to this account. The objective is to demonstrate how to analyze individual data to uncover similar opportunities.
One notable note on data limitations: historical Google Ads records show a Demand Gen middle-funnel campaign was disabled due to poor performance. Granular daily performance data for that failed campaign is unavailable because it was removed. Consequently, this examination contrasts that documented failure with the detailed, successful performance of the alternative strategy implemented on Meta to extract valuable insights.
Where Capital Goes: Quantifying the Middle-Funnel Leak
Every paid advertising funnel experiences leakage. Paying to send 1,000 visitors to a website yields an initial conversion rate on the first visit, while the remaining majority—often 95% or more—depart without converting. Visitors who show interest without converting form the middle-of-funnel audience. Retargeting exists to bring them back and secure conversion, yet most retargeting execution remains generic.
Within the SitterTree account, the top-of-funnel prospecting campaign on Meta targeting cold parent audiences acquired signups at $7.93 each, while Google parent signups averaged $5.56. Re-engaging someone who has previously visited the site and recognizes the brand should theoretically cost less as a warmer lead. However, the Google Ads log noted that the Demand Gen middle-funnel campaign was disabled due to poor performance, suggesting acquisition costs exceeded top-of-funnel levels or volume was negligible. Broad, generic video ads shown on YouTube to anyone who interacted with the brand lacked necessary nuance.
Treating a visitor who read three blog posts and subscribed to a newsletter the same as a visitor who bounced from the homepage within seconds generates massive waste. Capital spent at the top of the funnel loses value when follow-up engagement fails.
Quantifying this issue requires modeling against specific traffic numbers. A diagnostic tool below based on SitterTree funnel metrics allows input of monthly website visitors from ads to visualize drop-offs and test the impact of an efficient retargeting system.
Middle-of-Funnel Leakage Diagnostic
Pinpoint where value leaks from top-of-funnel spend using a baseline 24% click-to-signup rate from SitterTree Meta data.
The Alternative: Dissecting a High-Efficiency Middle-of-Funnel Engine on Meta
When broad automated approaches on Google underperform, successful middle-of-funnel campaigns rely on precision and segmentation rather than broad automation. The Meta retargeting campaign operated under a philosophy focused on structured audience subsets.
Top-line numbers indicate this retargeting campaign spent $1,572 and generated 450 completed registrations at a cost per registration of $3.50. Compared to the cold prospecting campaign cost of $7.93 on the same platform, retargeting acquired customers at less than half the price while outperforming top Google Ads campaigns for parents. Retargeting should serve as the most efficient conversion source.
Performance stemmed from ad set structure rather than random creative changes. The campaign segmented audiences by engagement intensity:
LH - US - Newsletter subscribers: Users who actively provided their email address, representing the highest-intent group.LH - US - Signups / Users: Users engaged in partial registration flows, representing a warm audience.LH - US - Page Engagers: Users who interacted with Meta pages through likes, comments, or shares.LH - US - Website visitors: The broadest category for general site visitors.
Segmenting audiences enabled intelligent budget allocation and messaging tailored to engagement depth. Data showed newsletter subscribers spent $1,084 and returned 318 registrations at $3.41 each. Page engagers produced lower costs at $2.64 per registration on a smaller scale, while general website visitors performed worst at $13.09 per registration. Lumping all visitors together overpays low-intent users while undervaluing high-intent prospects.
Granular segmentation aligns capital with user intent. The chart below illustrates cost differences across segmented audiences versus prospecting baselines.
Retargeting Audience Cost Efficiency vs. Prospecting
Cost Per Registration analysis from SitterTree Meta account.
Data Source: SitterTree Meta Ads performance breakdown.
The Compound Effect: How an Effective Middle-Funnel Strategy Lowers Overall Acquisition Cost
An efficient retargeting campaign impacts the entire marketing ecosystem by forming a feedback loop: top-of-funnel spend generates traffic and behavioral data, which powers retargeting sequences that lower blended cost per acquisition across the account.
SitterTree Meta performance data shows total spend of $11,388 yielded 1,663 registrations, resulting in a blended cost per acquisition of $6.85. Prospecting spent $9,615 for 1,212 registrations at $7.93 each, while retargeting spent $1,573 for 450 registrations at $3.50 each. Retargeting accounted for roughly 14% of spend while delivering over 27% of total conversions, dragging overall blended acquisition costs down.
Without retargeting efficiency, acquiring those extra 450 conversions through cold prospecting at $7.93 would have cost approximately $3,568 instead of $1,573—a direct saving of nearly $2,000. Efficient middle-funnel mechanics make total advertising spend generate higher overall return.
The simulator below illustrates how shifting budget allocations between top-of-funnel and retargeting influences blended acquisition costs.
Blended Acquisition Cost & Budget Allocation Simulator
Model budget shifts between prospecting and retargeting using SitterTree baseline figures.
Prospecting: $9,621 spend -> 1,213 conversions.
Retargeting: $1,566 spend -> 447 conversions.
From Analysis to Action: A Framework for Retargeting
Structured retargeting requires establishing a hierarchy of intent where messaging aligns with prior user actions and familiarity levels. Replicating success involves building distinct tiers rather than broad audience buckets.
The layered framework below reflects structures utilized across successful accounts, prioritizing high-intent audiences to capture immediate conversions.
The Intent-Layered Retargeting Framework
Layer 1: High Intent (The Almost-Converted)
Audiences: Newsletter Subscribers, Previous Customers, Started Checkout, Added Payment Info.
Strategy: Highest-value audience requiring direct calls to action and completion incentives, yielding low acquisition costs.
Layer 2: Medium Intent (The Engaged & Interested)
Audiences: Social Page Engagers, Video Viewers (50%+).
Strategy: Nurture interested users with testimonials, case studies, and distinct value propositions.
Layer 3: Low Intent (The Casual Visitor)
Audiences: All Website Visitors (excluding converters), Homepage Visitors.
Strategy: Broadest category utilized for brand reinforcement rather than direct hard selling.
Implementing this framework requires multi-stage nurture sequences with dedicated ad sets, specific budgets, and proper exclusions preventing audience overlap between layers.
While initial setup demands more effort than generic campaigns, structured retargeting avoids wasted capital and turns middle-funnel engagement into a profitable asset.
The Final Verdict: Stop Guessing, Start Segmenting
Evidence from the SitterTree account demonstrates that middle-funnel drop-offs stem from flawed strategies treating retargeting audiences as a monolithic group. Documented failures of broad Google Demand Gen campaigns contrasted against segmented Meta retargeting campaigns confirm the value of precision.
Relying on algorithms within generic retargeting campaigns creates wasted spend. A structured sequence of custom audiences built on user intent delivers higher conversions at lower costs and improves profitability.
Growth leaders should transition from generic retargeting to deliberate re-engagement sequences that respond to user signals. Identifying high-intent segments and prioritizing retargeting budgets plugs funnel leaks, lowers blended costs, and enables confident top-of-funnel scaling.
If you have looked at your own funnel and suspect a similar leak, the problem may lie in your strategy. Getting an expert review on your account can uncover structural issues costing capital daily.
For a detailed analysis of your ad accounts to identify and resolve funnel leaks, schedule a free, no-obligation strategy consultation. We can review your specific metrics and discuss a practical plan to improve performance.
Cite this Research & Data
If referencing this research, data, or strategy in your publication, please use the attribution below.
Lukas Holschuh
Founder, Growth & Advertising Consultant
Great campaigns fail without expertise. Lukas and his team provide the missing strategy, optimizing your entire advertising funnel—from ad creatives and copy to landing page design.
Backed by a proven track record across SaaS, eLearning, and eCommerce, they don't just run ads; they engineer systems that convert. A data-driven partnership focused on tangible revenue growth.