TLDR;
- You're running app install campaigns and wondering why the users are low quality compared to your web signups, despite the installs being cheap. The common assumption is that cheap volume is always good.
- I examined data from a real Google Ads account for an app. It shows app installs were 7.5 times cheaper than web registrations, but the web funnel still got over 77% of the total budget. This seems backwards, but it makes sense.
- The massive cost difference isn't a mistake; it’s a direct result of user intent, friction, and what you tell the ad platform's algorithm to do. App install campaigns are designed to find cheap downloaders, not committed users.
- Below, I've broken down the data, the mechanics behind this problem, and built a few interactive tools, including a calculator to model the true cost of low-quality users and a diagnostic matrix to help you choose the right strategy.
I see this question all the time. Founders and marketing managers get hooked on the low Cost Per Install (CPI) from their app campaigns. It looks great on a spreadsheet – maybe 30p, 50p, a pound per install. You feel like you're winning. But then you look at the actual usage numbers, the in-app purchases, the subscription rates... and they're quiet. Meanwhile, your web registration campaigns, driving traffic to a landing page, might cost you five or ten times as much per user, but those users stick around.
So, the question is, why? Why does the seemingly most efficient channel produce the lowest quality users? Is it just the price you pay for scale, something you just have to live with? Or is there a fundamental mechanical reason for it? I examined the data from a recent client campaign and it paints a clear picture. It's not about luck or bad creative; it's about the physics of the system. I decided to examine the numbers and lay out exactly what's going on under the hood.
My goal here is to go beyond the usual advice and get to the root cause, using real campaign data to show you why this happens and what it means for your strategy. This isn't just theory; it’s a close look at the mechanics of two very different user acquisition funnels.
First, what data are we actually looking at?
Before we get into the analysis, it helps to be clear about the data. This isn't a massive industry-wide study with thousands of advertisers. Most of those averages are useless anyway. This is data from a single, specific Google Ads account for a client of ours, Scout IQ. They have a utility app that lets people scan books to see their resale value online—a tool for people sourcing stock for their e-commerce businesses.
The dataset covers a specific performance period where we ran two primary types of campaigns simultaneously: one optimised for direct app installs on Android, and another series of campaigns (Search, Performance Max, Demand Gen) optimised for web-based account registrations. Both funnels led to the same outcome: a new user for Scout IQ. But they got there in very different ways, and at a very different cost.
The numbers I've pulled for this analysis are:
- App Install Campaigns: Generated 2,490 new users via direct installs at an average cost of £0.30 per install.
- Web Conversion Campaigns: Generated 1,168 new users via website registrations at an average cost of £2.25 per registration.
This single data set is clean for our investigation because it provides a side-by-side comparison of the two funnels operating under the same brand, targeting a similar overall audience, at the same time. Let's dig in.
The 7.5x Cost Gap: A Symptom, Not the Disease
The first thing that jumps out is the enormous difference in cost. At £0.30 per install versus £2.25 per web registration, the web-based user is 7.5 times more expensive to acquire. When you see a number like that, the knee-jerk reaction is to question the expensive channel. Why pay £2.25 for something you can get for 30p?
This is where many advertisers go wrong. They see cost as the primary metric of success and immediately want to shift all the budget to the cheaper channel. But the cost is just a symptom. It’s an indicator of what happens further down the line. It reflects the level of friction, intent, and qualification involved in each path. A higher cost per acquisition isn't a sign of failure; it’s a sign that you're acquiring a higher-quality asset.
User Acquisition Channel Performance
Direct comparison of top-level user acquisition metrics from the Scout IQ Google Ads account, splitting performance between the app install and web registration funnels.
To really hammer this home, look at the cost comparison visually. The difference isn't subtle; it's a chasm. This isn't a case of one channel being slightly less efficient. These are two fundamentally different acquisition models with wildly different economic outputs at the top of the funnel.
Cost Per User Benchmark: App vs. Web
Visualisation of the cost disparity between the two user acquisition funnels from the Scout IQ campaign data.
Source: Scout IQ Google Ads account performance data for the analysed period.
But where did the money actually go?
So if app installs are cheap, you'd assume we poured most of the budget into that campaign to maximise user volume, right? That’s the logical conclusion if you're only chasing a low top-line CPA. But the data shows the opposite. Let's calculate the spend for each channel.
- App Install Spend: 2,490 installs * £0.30/install = £747
- Web Conversion Spend: 1,168 registrations * £2.25/reg = £2,628
- Total Spend: £747 + £2,628 = £3,375
Despite being 7.5 times more expensive per user, the web registration campaigns received £2,628 of the budget, which is 77.9% of the total spend. The app install campaign only received £747, or 22.1%. This wasn't an accident. It was a deliberate strategic decision based on the expected downstream value of the users from each channel. We paid a premium for higher-intent users because they were more likely to become active, engaged customers. This decision challenges the conventional view that you should always scale the channel with the lowest CPA.
Campaign Budget Allocation by Funnel
Analysis of how the total ad spend was distributed between the high-cost web funnel and the low-cost app install funnel.
So, why the huge difference? It's about friction and intent.
Now we get to the core of the issue. The price difference isn't arbitrary. It's a direct reflection of the work a user has to do to convert, and what that work says about their level of interest. Think about the two journeys.
The App Install Journey (Low Friction): A user scrolls through YouTube, their Gmail inbox, or browses the Google Play Store. They see an ad. It looks interesting. They tap 'Install'. The app downloads in the background. The entire process takes one or two taps and a few seconds. There's almost no commitment. It's an impulse action. They might open the app once and forget about it, or never open it at all. The barrier to entry is practically zero.
The Web Registration Journey (High Friction): A user sees an ad, maybe after searching for something like "book selling app" on Google. They click the ad, which takes them to a landing page. They read the copy, understand the value proposition, and decide this tool is worth their time. Then they find the sign-up form, type in their name, email address, create a password, and click 'Submit'. This multi-step process requires effort, time, and cognitive load. It acts as a natural filter. Anyone who completes this journey demonstrates a higher level of intent. They aren't just curious; they're actively seeking a solution.
Every extra step a user takes sheds lower-intent people from the funnel. The web funnel has more steps, so only the motivated make it through. You pay more for them because they are rarer and more valuable. The app install funnel removes all friction, which lets everyone in, motivated or not.
User Funnel Architecture: App vs. Web
Low-Friction App Install Funnel
High-Friction Web Conversion Funnel
Are you telling the algorithm to find customers or just downloaders?
The second major piece of the puzzle is the campaign objective you choose. This is an overlooked aspect of campaign strategy. When you create a campaign in Google Ads or Meta, the first thing it asks is your objective. Your answer tells the platform's machine learning algorithm what success looks like and who to target.
If you choose "App promotion" with a goal of "App installs," you give the algorithm a direct command: "Find the people within my targeting parameters who are most likely to click a button and download an app, for the lowest possible cost." The algorithm excels at this. It scours data to find users who frequently download apps impulsively and cost little to reach. It ignores whether these users ever open the app, register, or spend money. You asked for installs, and it delivers installs, cheaply and in volume.
If you choose "Leads" or "Sales" with a conversion goal of a "Website Registration," you give it a harder, more specific task: "Find people who are not only likely to click but also likely to navigate a website, fill out a form with their personal details, and complete a registration." The algorithm must search for higher intent signals: people who fill out forms, spend time reading content, and research solutions. These users are more valuable and in higher demand, so they cost more to reach. But they are pre-qualified by the algorithm to complete a high-friction action.
You get what you optimise for. If you tell Google you want cheap installs, it finds people who install apps. If you tell it you want registered users, it finds people likely to become registered users. The lower quality is a direct consequence of the instructions you give the machine.
Let's model the true cost of low quality users
This is where it all comes together. A low CPI is a surface-level metric if those users never convert into active, registered customers. The real metric you should care about is the Cost Per Valuable Action—in this case, an actual account registration. For the web funnel, the Cost Per Install matches the Cost Per Registration (£2.25), since the registration is the conversion event.
But for the app install funnel, there's a second, hidden drop-off point: the percentage of people who install the app and register an account inside it. This represents user quality. If 50% of installers register, your true cost per registration is double your CPI. If only 10% register, your true cost is ten times your CPI.
Let's use a calculator to see how this plays out. We'll plug in our known costs and volumes and add a slider for the In-App Registration Rate. Watch what happens to your Effective Cost Per Registered User from the app channel as that quality rate drops.
The True Cost of Low-Quality App Installs
This tool models how the post-install registration rate impacts the actual cost to acquire a registered user from an app install campaign, compared to the web funnel.
Try sliding the In-App Registration Rate down to 13%. At that point, your effective Cost Per Registration from the app channel becomes £2.30, which is now more expensive than the web funnel. Anything below that, and your cheap app install campaign is actually a more expensive way to get registered users. Given that for many utility apps, a post-install registration rate of 5-10% is common, you can see how quickly the economics fall apart. The cheap CPI is a mirage; the only thing that matters is the final cost per quality action.
So how do you choose the right path for your app?
This doesn't mean app install campaigns are useless. They have their place. They are a tool for driving mass awareness and top-of-funnel volume cheaply. If your app has broad appeal, a simple value proposition, and monetises through ads or small in-app purchases, a volume-focused strategy works. You're playing a numbers game.
If your app requires more commitment—a subscription, a significant purchase, or is a B2B tool like Scout IQ—then user intent matters more than raw volume. You need users actively looking for a solution and willing to invest time to learn your product. In these cases, a higher-friction web funnel, despite its higher top-line CPA, is more profitable long-term.
The key is understanding trade-offs and using the right tool for the job. Here's a simple matrix to help you diagnose which approach fits you best.
Funnel Strategy Diagnostic Matrix
A framework for deciding between an App Install and Web Conversion strategy based on campaign goals and user characteristics.
| Factor | App Install Funnel | Web Conversion Funnel |
|---|---|---|
| Primary Goal | Volume, Brand Awareness, Top of Funnel | Lead Quality, High-Intent Users, Sales |
| User Intent | Low (Passive, Impulse-driven) | High (Active, Problem-solving) |
| Friction | Very Low (1-2 taps to install) | High (Site visit, form fill, etc.) |
| Algorithm Objective | Finds cheapest impressions likely to install | Finds users likely to complete a specific action |
| Typical Top-Line CPA | Very Low (£0.20 - £2.00) | High (£2.00 - £50.00+) |
| Best For... | Gaming apps, mass-market social/utility, ad-monetised apps | SaaS, B2B, eCommerce, high-consideration services, subscription apps |
The bottom line: you get what you pay for
The mystery of low-quality app installs isn't a mystery. It's economics and system design. The low cost per install isn't a discount; it's the market price for a low-intent, low-friction action performed by a user the algorithm identifies as cheap to reach, not valuable to own.
Our analysis of the Scout IQ account data proves this. The web funnel, despite being 7.5x more expensive at the top, was the strategic priority, receiving over 77% of the budget. This is the decision you make when you stop obsessing over top-line CPA and focus on the entire funnel, user quality, and lifetime value. A cheap lead that never converts is more expensive than a qualified lead that becomes a long-term customer.
When you see that temptingly low CPI in your app install campaign, don't take it at face value. Ask what you're really buying. Are you buying a customer, or just a download?
Building a strategy balancing volume and quality requires understanding these mechanics. It's about knowing when to use a cheap install campaign and when to deploy a high-intent web funnel. If you're struggling to find that balance in your campaigns and are tired of paying for installs that go nowhere, consider an expert review. We offer a free, no-obligation strategy consultation where we look at your data together and figure out the right approach 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.