What are the main takeaways from this analysis?
- Your overall account CPA looks healthy, but you suspect your actual cost to acquire a real customer is much higher. You're probably right; blended metrics hide serious budget leaks.
- The data from this childcare app account shows that high-volume, low-cost signups (job-seeking babysitters) were masking a dangerously high cost to acquire actual customers (parents). The blended account CPA of $2.98 hid a true Parent CPA of over $10.00 in the struggling search campaign.
- The leak was caused by ambiguous, high-intent keywords like "babysitter needed" attracting both parents and job seekers. Without strict negative keyword lists, up to 50% of the budget for these terms was being wasted on non-customer clicks.
- Below, you'll find an interactive calculator to diagnose this exact problem in your own campaigns, plus a breakdown of the strategy used to cut the Parent Search CPA and stabilise the account.
I get asked this question a lot by founders and growth leads: "Lukas, my overall account CPA is right on target, but my sales team is complaining about lead quality and we just don't seem to be growing. What gives?" Nine times out of ten, the answer is hiding in plain sight, obscured by a misleading average. You're looking at a blended cost per acquisition that's being artificially dragged down by low-cost, low-value conversions, while the cost to acquire an actual, paying customer is spiralling out of control.
This isn't just a theory. It's a common and expensive problem that I see in accounts almost every week. So, my research question for this piece is simple: How much budget is really being wasted on the wrong audience in a search campaign, even when the top-line metrics look healthy? And what specific, repeatable tactics can be used to diagnose and fix the leak?
To get to the bottom of this, I'm going to pull apart a real Google Ads account for a childcare services app. We'll examine how an ambiguous search term can attract two completely different audiences—parents looking to hire and sitters looking for a job—and quantify the exact financial damage this intent bleed causes. More importantly, I'll show you the exact structural changes that brought the costs back in line.
Before we get into the numbers, a quick bit of housekeeping on the data itself. Everything we're looking at comes from a single SitterTree account, covering performance across Google Ads and Meta Ads from roughly July 2024 to late November 2024, with some more recent snapshots. In total, the Google Ads data covers around $13,500 in spend and 4,500 conversions. This isn't an industry-wide study; it's an investigation into one account's journey. The patterns are specific, but the principles for fixing them are universal. Don't treat these numbers as benchmarks for your own business, but as evidence for a strategic approach.
What's really going on with the account's CPA?
At first glance, the Google Ads account looks to be in great shape. The all-time performance summary shows a total of 4,529 conversions at an average Cost Per Conversion of just $2.98. If the goal was simply to get signups for under $3.00, you'd probably give yourself a pat on the back and crack on. But this is the classic trap of the blended metric.
The business has two distinct audiences: "Sitters" (people looking for babysitting jobs) and "Parents" (people looking to hire a babysitter). The founder was clear that getting Sitters is easy and cheap; the real challenge and business goal is acquiring Parents. When we segment the performance, the illusion of a low CPA shatters immediately.
Let's do the maths. The account acquired 3,085 Sitters at an average cost of $1.78 each. It also acquired 1,444 Parents, but at an average cost of $5.56 each. The high volume of cheap Sitter signups (over double the number of Parent signups) completely masks the much higher cost of acquiring an actual customer.
(3,085 Sitter signups * $1.78/signup) + (1,444 Parent signups * $5.56/signup) = $5,491 + $8,029 = $13,520 total spend.
$13,520 spend / 4,529 total signups = $2.98 blended CPA.
The account average isn't just misleading; it's irrelevant to the core business goal of acquiring parents cost-effectively. The number that actually matters, the Parent CPA of $5.56, is almost double the account average. And as we'll see, within the campaigns specifically designed to find Parents, the problem was even worse.
The Blended CPA Illusion
Metric audit showing how low-cost, high-volume Sitter acquisitions artificially deflate the account's average CPA, hiding the true, much higher cost of acquiring a Parent.
Where is the parent acquisition budget actually leaking?
Knowing the overall Parent CPA is high isn't enough; we need to find the specific point of failure. The account contains multiple campaigns targeting parents, including Performance Max and Search. While the PMax campaign for Parents performed surprisingly well (888 conversions at $2.25), the dedicated Search campaign was a different story. The data shows the `QLH - Search (Parents)` campaign had an all-time CPA of $10.00, and a more recent snapshot put it at $7.83. This is where the real budget haemorrhaging was happening.
A well-structured search campaign should be a precision tool. By splitting keywords into tightly themed ad groups based on user intent—from early-stage research to ready-to-buy actions—you control the message and the budget. The structure in this account followed that practice, with ad groups for different stages of the buyer journey: `Upper Funnel: Info`, `Middle of Funnel: Cost`, and `Lower Funnel: Hiring`, `Near Me`, and `Competitors`.
In theory, the `Lower Funnel: Hiring` ad group, targeting people actively looking to employ a babysitter, should be one of the most efficient. These are users with the highest commercial intent. But when we look at the numbers, that's not what happened at all.
Parent Search CPA by Ad Group Theme
Cost per acquisition (CPA) for Parent signups, broken down by the keyword theme of each ad group within the main Search campaign. This isolates the source of inefficiency.
The `Lower Funnel: Hiring` ad group had a CPA of $10.53, making it one of the worst-performing segments in the entire campaign. It was even more expensive than the `Upper Funnel: Info` group ($10.14), which is targeting people who are just starting their research. This is completely backwards. High-intent, bottom-of-funnel keywords should convert better and at a lower cost, not worse. This data shows that something is fundamentally wrong with the traffic being attracted by these hiring keywords.
What's making the 'Hiring' ad group so inefficient?
To find the culprit, we have to go one level deeper and look at the specific search keywords that triggered the ads in this ad group. This is where the root of the problem becomes clear. The campaign is trying to find parents, but the keywords are also perfect for attracting job-seeking babysitters.
Consider the top-performing keyword by volume in this ad group: `"babysitter needed"`. On the surface, it seems perfect. A parent looking for childcare would search exactly that. But so would a student looking for a part-time job. They might search "jobs for when a babysitter needed" or "I am a babysitter needed by a family". Because the keyword is set to Phrase Match, it can trigger for a wide range of searches that contain that core phrase. The intent is completely ambiguous.
'Lower Funnel: Hiring' Keyword Breakdown
Performance metrics for the top keywords within the highest-cost ad group. Note the unusually high CTR for `"babysitter needed"`, a strong indicator of mixed, low-quality traffic.
| Keyword | Conversions | Cost / Conv. | CTR | Total Cost |
|---|---|---|---|---|
| "babysitter needed" | 102.47 | $10.41 | 15.40% | $1,067.04 |
| "nanny needed" | 34.88 | $6.76 | 8.03% | $235.78 |
| "looking for a babysitter" | 32.57 | $12.94 | 7.78% | $421.35 |
| "hire a babysitter" | 3.62 | $20.01 | 3.66% | $72.30 |
The data proves this out. The keyword `"babysitter needed"` has a click-through rate (CTR) of 15.40%. For a commercial search campaign, that is exceptionally high. It suggests the ad is appealing to a much broader audience than just parents. Job seekers, who are looking for opportunities, are far more likely to click on anything that looks like a job lead. Compare that to a more specific, less ambiguous keyword like `"hire a babysitter"`. Its CTR is only 3.66%—much more realistic for a targeted commercial query—but its CPA is a shocking $20.01. The campaign is paying a premium for clicks from the wrong audience on its highest-volume keyword.
This is the definition of a budget leak. Every click from a job seeker on an ad intended for a parent is wasted ad spend. It inflates clicks, impressions, and costs, but it will never generate a Parent conversion. This drives the CPA through the roof. We can model this financial damage with a few simple calculations.
The Intent Mismatch Cost Calculator
See how clicks from the wrong audience (e.g., job seekers) on your customer acquisition campaigns inflate your true CPA. Adjust the sliders based on your own campaign data.
How do you systematically fix this kind of intent bleed?
Diagnosing the leak is one thing; plugging it is another. The solution isn't to simply pause the awkward keywords. Terms like `"babysitter needed"` are high-volume and high-intent—you can't afford to ignore them. The fix requires a two-pronged structural approach: refining your ad group themes and building an aggressive negative keyword shield.
First, you must enforce a strict Single-Theme Ad Group (STAG) structure. This means moving beyond broad journey stages like "Lower Funnel" and creating ad groups around extremely tight keyword concepts. For example, keywords related to *needing* care should be in a separate ad group from keywords related to *hiring* care. While they seem similar, the language reveals a subtle but important difference in intent that Google's algorithm will pick up on.
Second, and most importantly, you have to be rigorous with your negative keywords. For this account, the most helpful action was creating and applying a comprehensive "Jobs & Careers" negative keyword list to every single Parent-focused campaign. This list acts as a firewall, explicitly telling Google: "If the search query contains any of these words, do not show my ad." The transcript mentions excluding terms related to "gigs" and "jobs," but a robust list would go much further, including words like: application, apply, career, employment, hiring, resume, CV, salary, wage, vacancy, work, and so on.
This combination of tight ad groups and a strong negative list transforms your campaign from a wide net into a spear. You're no longer hoping the right person sees your ad; you're actively filtering out everyone who isn't your ideal customer before you spend a single penny on their click.
The Intent Filtering Framework
Layer 1: The Campaign-Level Negative Shield (Mandatory)
Apply a master "Jobs & Careers" negative keyword list to all Parent-focused campaigns. This is your first line of defence to block any search containing terms like job, career, salary, apply, gig, vacancy.
Layer 2: Single-Theme Ad Group (STAG) Structure
Separate keywords into ultra-specific themes. Don't mix "babysitter needed" with "hire a babysitter". This allows for more relevant ad copy and better control, forcing ambiguous terms into their own sandboxed ad group for monitoring.
Layer 3: Search Query Report Pruning (Regularly)
Review the actual search terms report for your ambiguous keywords. Add any new, irrelevant job-seeking variations as ad-group-level or campaign-level negatives. This is an ongoing process of refinement.
Can we actually see the impact of these fixes in the data?
This is the most satisfying part of the investigation. It's one thing to have a strategy; it's another to see it deliver tangible results in the account's month-on-month performance data. The data clearly shows a dramatic improvement in the Google Ads CPA right around the time these structural optimisations would have been implemented.
In the initial launch month (July-August 2024), the blended Google Ads CPA was $4.91. This was the period of initial learning and testing, where the intent bleed from job seekers was likely at its worst. After the first round of optimisations, which included pausing a poorly performing Demand Gen campaign, the CPA for the next month (August-September) dropped significantly to $2.64. Then, following the period noted for optimising search keywords and negative lists (September-October), the CPA stabilised at an even better $2.54. That's a 48% reduction in acquisition cost from the starting point.
Google Ads CPA Compression Curve (4-Month View)
The measured blended CPA reduction across the Google Ads account following campaign restructuring, keyword refinement, and the implementation of a strict negative keyword strategy.
This isn't a coincidence. By systematically choking off the supply of irrelevant job-seeker traffic to the high-intent Parent campaigns, the budget was automatically reallocated to more relevant, higher-quality clicks. The result is a lower cost per click from the right audience, a higher conversion rate, and a dramatically lower cost per acquisition. The campaign stopped wasting money and started working efficiently, ultimately helping bring the all-time account CPA down to the target of ~$3.00.
What does this mean for your own campaigns?
If there's one thing you take away from this analysis, it should be this: never trust a blended CPA. Averages hide both your biggest wins and your most significant budget leaks. You have to segment your data by your true business goals. If you have two different audiences, you need to track their acquisition costs separately. No exceptions.
Second, be sceptical of any keyword with an unusually high CTR, especially for bottom-of-funnel campaigns. It's often not a sign of a great ad, but a symptom of a keyword that's too broad or ambiguous, attracting a low-quality, click-happy audience that will never convert. Go straight to your search terms report and see what people are actually typing. You'll likely find the source of the problem there.
Finally, treat your negative keyword lists not as a set-and-forget task, but as your primary tool for budget defence. A well-maintained set of negative lists is the single most powerful lever you have for improving efficiency in a Google Search account. It's the boring, unglamorous work that separates the amateurs from the professionals who get real results.
This deep dive was based on one account in a specific niche, so your mileage may vary. But the principles of intent mismatch, budget leakage, and surgical optimisation are universal. The same forces are at play in your B2B SaaS campaign targeting "project management software" (and getting clicks from students looking for free tools for a school project) or your e-commerce campaign for "leather work boots" (and getting clicks from people looking for boot repair services).
Getting this right isn't about finding a magic trick. It's about a relentless, thorough approach to the data and a disciplined campaign structure. It takes time and expertise to untangle these issues, especially in a mature account. Many teams simply don't have the bandwidth to do this level of deep-dive analysis while also managing creatives, landing pages, and overall strategy.
If you've looked at your own metrics while reading this and suspect you have a similar leak hiding in your account, it might be time for a second opinion. We offer a free, no-obligation strategy consultation where we can crunch your metrics with you and figure out what might be holding you back. It's often a few small, structural changes that can unlock significant growth and make your ad spend work a lot harder.
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.