Overview
- Two-sided marketplaces like childcare apps often struggle with one side (parents) being drastically more expensive to acquire than the other (sitters), leading to unpredictable growth. The common assumption is that this is an unavoidable cost of business.
- Analysis of ad performance for one such app shows this asymmetry is not just a market problem, but a channel and targeting problem. Parent acquisition costs swung from $2.25 to over $13.57 depending on the platform, campaign type, and audience used.
- The data proves that a blended strategy, combining high-intent Google Search with precisely layered Meta prospecting and aggressive retargeting, is what stabilises costs and allows for scale. The most efficient cold traffic didn't come from obvious working parent targets, but from layering demographics with niche interests like Baby Essentials.
- Below, I've broken down the data and built a few interactive tools, including a channel performance database and a scenario simulator, to show how you can model a more predictable acquisition engine for your own marketplace.
For any founder running a two-sided marketplace, there's a familiar, nagging asymmetry. One side of the market signs up with little effort, while the other feels like you're pushing water uphill with a fork. In the childcare space, it's a classic: getting babysitters, the supply side, is cheap and easy. Getting parents, the demand side, to sign up is a different beast entirely—expensive, unpredictable, and a constant brake on growth. The common wisdom is to just accept it. Parents are busy, harder to reach, so you pay the premium. But is that the full story?
That led me to a pretty specific research question: Can the wild, unpredictable swings in parent acquisition costs be tamed into a scalable, predictable model by systematically segmenting channels and audiences, or is it just an unavoidable cost of doing business?
To examine this, I'm not using some broad industry survey. I'm going deep on one specific dataset: the ad accounts for SitterTree, a childcare services app. The data covers performance from mid-2024 to early 2025 across their Google and Meta advertising, totalling over 5,700 signups. By treating this single account as a controlled environment, we can isolate what’s actually driving performance and what’s just noise. This isn't a case study of what we did; it's an investigation into what the numbers actually reveal about solving this lopsided growth problem.
First, just how bad is the acquisition gap?
Before getting into channels and funnels, you have to appreciate the scale of the problem. When you look at the top-line numbers for SitterTree, the imbalance is stark. Over the core reporting period (July to November 2024), the supply side was incredibly efficient, while the demand side was, to put it mildly, a lot more work.
Let’s break it down. For sitters, the cost per conversion on Google Ads was a tidy $1.78. Some campaigns, like the Performance Max Sitter campaign, were even getting signups for as little as $1.22. It’s the kind of metric that makes a founder sleep well at night. On the other side of the coin, parents, the blended cost per conversion on Google was $5.56. On Meta, it was even higher at an average of $7.07. That’s a 3-4x difference right out of the gate. And as we’ll see later, those parent acquisition costs had a nasty habit of spiking much, much higher.
This isn't just a small variance; it's a fundamental business challenge. You can't scale a marketplace if one side costs four times as much to acquire and is prone to unpredictable cost explosions. It creates a constant state of firefighting rather than strategic growth. The chart below puts these core metrics side-by-side.
Marketplace Acquisition Asymmetry
Executive metric summary of the core acquisition cost difference between the supply (Sitters) and demand (Parents) sides of the marketplace.
So, is one channel just better at finding parents?
When you have a big cost difference, the first question is always about channels. Is Google just naturally better than Meta for this, or vice versa? A knee-jerk reaction would be to find the cheaper channel and pour all the money into it. But the data shows it’s not that simple. Each channel has its own flavour of efficiency and its own pitfalls.
Let’s look at the parent acquisition campaigns specifically. On Google, the blended parent CPA was $5.56. Within that, though, there were huge differences. The Performance Max campaign for parents was a standout, bringing in 888 signups at an average cost of just $2.25. That's almost as cheap as acquiring sitters. On the other hand, the Google Search campaign for parents was significantly more expensive, with a CPA closer to $10.00 according to the ad group data.
Over on Meta, the story is different again. The overall average cost per registration was $7.07. Prospecting campaigns averaged $7.93, while retargeting was a very efficient $3.50. So, Meta was great at converting warm audiences but more expensive than Google Performance Max for finding cold ones. The chart below lays out this channel-by-channel performance. It becomes pretty clear that there's no single best channel; instead, there are specific campaign types within each channel that are doing the heavy lifting.
Parent Acquisition Cost by Campaign Type
A comparison of the Cost Per Signup for parents across different campaign types on Google and Meta. The blended average across all parent campaigns is $6.21.
Is high-intent search really the answer?
Seeing Google Search as the most expensive channel is a bit of a head-scratcher. The usual advice is that search captures high-intent users who are actively looking for a solution, so they should be cheaper to convert. But here, with a CPA of around $10, it’s the least efficient channel for acquiring parents. So what’s going on?
The strategy for the parent search campaign was to split ad groups by the user's journey: Upper Funnel (informational searches like how to find a babysitter), Middle Funnel (cost-related searches), and Lower Funnel (hiring-intent searches like babysitter needed). You’d expect the Lower Funnel hiring group to be the most efficient. It has the highest intent, after all.
But the data tells a more complicated story. The Upper Funnel Info ad group actually generated the most conversions (243) at a CPA of $10.14. The Lower Funnel Hiring group was right behind it with 184 conversions, but at a slightly higher CPA of $10.53. The other lower and middle funnel groups had far less volume. This suggests that for a service like childcare, parents start with broad, informational questions and can be converted from there just as effectively—if not more so—than when they use ready to buy keywords. It challenges the classic marketing funnel dogma that says you should only focus on bottom-of-funnel keywords for conversions. In this market, the top of the funnel is just as valuable.
To let you explore this yourself, I’ve put the ad group data into a searchable table below. You can filter by funnel stage to see how the metrics stack up.
Google Parent Search: Ad Group Performance
| Ad Group | Funnel Stage | Cost / Conv. | Conversions | CTR | Total Cost |
|---|---|---|---|---|---|
| Upper Funnel: Info | Upper | $10.14 | 243.09 | 5.72% | $2,466.05 |
| Lower Funnel: Hiring | Lower | $10.53 | 183.75 | 10.26% | $1,935.46 |
| Lower Funnel: Near Me | Lower | $11.42 | 54.03 | 6.21% | $617.11 |
| Lower Funnel: Competitors | Lower | $7.23 | 37.36 | 2.96% | $269.95 |
| Upper Funnel: Providers | Upper | $7.12 | 35.91 | 7.60% | $255.61 |
| Middle of Funnel: Cost | Middle | $8.93 | 1.93 | 2.78% | $17.26 |
If not intent, what about audience layering on Meta?
If high-intent search isn't the magic bullet, then maybe the answer lies in Meta's powerful audience targeting. The challenge on Meta isn't intent; it's finding pockets of relevance at scale. How do you find parents of young children without just targeting everyone and burning through your budget?
The strategy here was to start with a core demographic, Parents (0-5 years), and then layer on additional interests or behaviours to refine the audience. This is where things get really interesting. The common-sense approach would be to layer on something like Working parent, assuming they're the ones most in need of childcare. But that was one of the worst-performing audiences, with a Cost Per Registration of $9.68.
The standout winner, by a long shot, was an audience that layered Parents (0-5 years) with an interest in Baby Essentials. This group brought in 401 registrations at a CPR of just $6.76. That's 30% cheaper than the Working parent audience. Why? My theory is that an interest in Baby Essentials is a much stronger, more immediate signal of someone's life stage than a broader behaviour like working parent. It captures parents who are actively in the thick of it—buying nappies, researching prams, and feeling the daily pressures that lead to needing a babysitter. It’s a proxy for urgency. Broad targeting (just parents 0-5) also performed well at $7.05, suggesting that once the Meta pixel has enough data, it can find the right people without heavy-handed layering. The chart below compares the performance of these different audience layers. It's a clear lesson in finding the right signals of intent, even on a platform that isn't intent-driven.
Meta Prospecting Audience Performance (Cost Per Reg)
Cost Per Registration for different Top-of-Funnel audience layers on Meta. The base demographic for all is Parents (0-5 years).
Data source: SitterTree Meta Ads account performance, TOFU campaign ad sets.
How do you build a predictable model out of all this?
So we've established that some channels and audiences work better than others. Google Performance Max is great for cheap, cold reach. Google Search captures intent, but at a higher cost than you'd think. Meta is brilliant for retargeting and can be effective for prospecting if you find the right audience layer. But how do you combine these into a system that doesn't just work, but is predictable and scalable?
The central task is to stop thinking in terms of a single best channel and start thinking in terms of a portfolio. You need a mix of strategies that work together. This is also how you protect yourself from the volatility we saw in the data, where the Meta CPR spiked to over $13 in March 2025 after a budget cut. Relying on one campaign is a recipe for disaster.
To make this tangible, I've built a scenario simulator below. It lets you model out a budget allocation based on the real performance data from the SitterTree account. You can choose a baseline campaign to build from, set a target for monthly signups, and then toggle different strategic levers—like activating retargeting or shifting budget to a better-performing channel—to see how it impacts your overall spend and blended CPA. This turns a complex set of data points into a practical tool for planning and forecasting. Play around with it. You'll see that the most stable, cost-effective models come from a blend, not from putting all your eggs in one basket.
Parent Acquisition Cost & Strategy Simulator
Select a starting channel baseline, set your monthly signup goal, and toggle the strategic levers to see the projected impact on your blended CPA and budget.
Pulling it all together: A blueprint for taming asymmetry
After digging through all this data, a clear picture emerges. The wild swings in parent acquisition cost aren't just random market noise. They are a direct result of over-reliance on a single channel, campaign type, or audience. Predictability doesn't come from finding one silver bullet, but from building a layered system where each component plays a specific role.
Based on this research, a robust acquisition engine for a childcare marketplace should look something like this. It’s a multi-layered framework designed to capture demand at every stage of awareness and intent, balancing cost and volume.
A Predictable Parent Acquisition Framework
Layer 1: Efficient Cold Prospecting (The Volume Engine)
Channels: Google Performance Max & Meta Prospecting. Job: To generate the largest volume of new, relevant eyeballs at the lowest possible cost. PMax was the winner here at $2.25/signup. Meta's Parents (0-5) x Baby Essentials audience at $6.76 provides scalable reach on social.Layer 2: High-Intent Capture (The Hand-Raisers)
Channel: Google Search. Job: To catch parents actively searching for a solution. Focus on keywords like babysitter needed but don't ignore informational queries. The data showed these convert well. This layer is more expensive ($10 CPA) but provides a steady stream of high-quality leads.Layer 3: Aggressive Retargeting (The Safety Net)
Channel: Meta Retargeting. Job: To convert the 95%+ of people who visit the site from other channels but don't sign up immediately. With a CPA of just $3.50, this is the most efficient part of the entire system and is non-negotiable for maximising budget.The problem of marketplace asymmetry, it turns out, isn't something you solve once. It's something you manage continuously with a diversified portfolio of campaigns. The data from SitterTree shows that while sitters will always be cheaper to acquire, a thoughtful, data-driven approach can make parent acquisition far more predictable and efficient than just throwing money at the wall and hoping for the best.
It requires moving beyond surface-level metrics and digging into which specific campaigns, ad groups, and audiences are actually driving results. It's a lot of work, and it's easy to get lost in the weeds. This kind of forensic analysis and system-building is exactly where having an expert partner can make all the difference, turning a volatile cost centre into a predictable growth engine.
If you're wrestling with a similar problem in your own business and feel like you're just guessing, it might be time for a fresh pair of eyes. We offer a free, no-obligation strategy consultation where we can crunch your metrics and figure out what a more predictable acquisition system could look like for you.
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.