TLDR;
- Our Strategic Fix: We built a multi-channel Google Ads system, testing App Installs, Search, Performance Max, and Demand Gen campaigns simultaneously to identify the most efficient channels for user acquisition.
- Verified Results: Drove 3,543 total new users (2,482 app installs and 1,061 web registrations) at a blended average cost of just £0.96 per user.
- Key Takeaway: Don't bet on a single Google Ads campaign type. A portfolio approach lets you find the cheapest user acquisition channels, then you can bin the losers and pour budget into what actually works.
Most people who come to us with a Google Ads problem are making the same mistake. They’ve set up a single Search campaign, thrown some keywords at it that they *think* their customers are using, and are now wondering why they’re burning through cash with nothing to show for it. They treat Google like a one-trick pony, and it’s costing them a fortune. The truth is, the Google ecosystem is a sprawling network of different platforms, and if you’re only using one bit of it, you’re missing the point.
This is a story about how we took a more structured approach for a client, ScoutIQ. They have an app that helps people who sell books online, mainly on Amazon. You scan a book's barcode, and the app tells you what it's worth, so you know if it's profitable to flip. It's a smart tool for a very specific audience. Instead of just guessing what might work, we built a system to test multiple parts of Google’s network at once. The result? We brought in 3,543 new users for them at an average cost of just £0.96 each. That’s less than a quid for a new potential customer. Here’s how we did it.
The Problem with Sticking to What You Know
Before we ever build a campaign, we map out a proper strategy. It’s not the exciting bit, but it’s the bit that stops you from wasting money. We did our homework: competitor research, digging into keywords, and outlining exactly what campaigns we were going to test and how we would optimise them. Too many people skip this and jump straight into the ads manager, which is like trying to build a house without a blueprint. It's bound to fall apart.
The standard advice you'll see online is to "start with Search ads". It makes sense on the surface. People are actively searching for a solution, so you show them an ad. Job done. But what about all the people who have the problem but don't know a solution like yours even exists? What about the people browsing the app store, or watching YouTube, or checking their email? You're ignoring them completely.
Relying only on Search means you're fighting with every other competitor over a very small pool of high-intent people. Costs go up, and you hit a ceiling fast. We knew that to get the volume and low cost ScoutIQ needed, we had to think bigger. We had to treat Google Ads not as a single tool, but as a portfolio of different channels, each with its own job. The plan wasn't to find the one "perfect" campaign type, but to run several in parallel and let the data tell us where to put the money.
This is a completely different mindset. It's not about finding what works and sticking to it. It's about building a system that is constantly testing, learning, and reallocating budget to the most efficient parts of the machine. It’s an active process, not a "set it and forget it" job.
Our Initial Setup: A Multi-Pronged Attack on Google
So, we mapped out a multi-channel test. We wanted to hit potential customers at every possible touchpoint within the Google universe. This meant setting up four distinct campaign types, each designed to do a different job.
First up was the App Installs campaign. This was a no-brainer. ScoutIQ has an Android app, and this campaign type is designed specifically to get people to install it directly from the Google Play Store. These ads show up when people are searching the store or browsing for other apps. The big appeal here is usually volume. You can often get a huge number of installs for a very low cost. The catch, as we’ll get into, is that the quality of these users can be a bit dodgy. The campaign optimises for the install itself, not for what the user does afterwards. But as a way to fill the top of the funnel quickly and cheaply, it’s often worth a punt.
Next, we built out a classic Search campaign. This was our tool for capturing "hand-raisers" – the people who are actively aware of their problem and are searching for a fix. We're talking about keywords like "book scanner app" or "how to sell books on Amazon". This is where you find your most motivated potential customers, but it's also the most competitive space. Our angle here wasn't just to target obvious keywords, but to structure the campaign around the buyer's journey, splitting our ad groups into different levels of intent, from broad awareness to people searching for direct competitors.
Then we added a Performance Max (PMax) campaign. PMax is Google's all-in-one campaign type that runs across every single one of their properties: YouTube, Display, Search, Discover, Gmail, and Maps. It’s a bit of a "black box" – you feed it your assets (headlines, images, descriptions) and your conversion goal, and Google's AI takes care of the rest. A lot of advertisers don't like it because you lose a lot of control. But we see it as a powerful discovery tool. We used it to see if Google’s algorithm could find pockets of converting users in places we might not have thought to look, at a cost that made sense.
Finally, we tested a Demand Gen campaign (what used to be called Discovery). These ads appear in Google's feeds, like the YouTube home feed, the Gmail promotions tab, and the Google Discover feed on mobile. We set this up with two goals. One was for top-of-funnel (TOFU) prospecting, trying to generate awareness among people who fit the target audience profile but weren't actively searching. The other was for middle-of-funnel (MOFU) retargeting, showing ads to people who had already visited the ScoutIQ website but hadn't signed up. The idea was to stay top-of-mind and nudge them towards converting.
With this four-pronged setup, we were covering all the bases: active searchers, app store browsers, passive feed-scrollers, and previous website visitors. Now it was time to launch and see which channels would actually deliver.
Initial Campaign Performance Test
This table breaks down the performance of each campaign type we tested. The goal was to identify which channels could acquire new users—either app installs or web registrations—at a profitable cost.
| Campaign Type | Purpose | Total Cost | Total Users Acquired | Average Cost Per User |
|---|---|---|---|---|
| App Installs - Android | Drive high-volume installs from the Play Store | £754.90 | 2,490 | £0.30 |
| Performance Max | Find users across all Google channels (AI-driven) | £665.89 | 554 | £1.20 |
| Search Ads | Capture high-intent users actively searching | £1,538.32 | 592 | £2.60 |
| Demand Gen (TOFU) | Awareness for cold audiences in feeds | £287.54 | 16 | £17.97 |
| Demand Gen (MOFU) | Retargeting previous visitors in feeds | £140.69 | 6 | £23.45 |
Letting the Data Talk: Finding the Winners and Binning the Losers
After letting the campaigns run and gather enough data, the picture became incredibly clear. As you can see from the table above, there was a massive difference in performance between the campaign types. This is the moment of truth in any campaign, where you have to be brutally honest and follow the numbers, not your gut feeling or what you *hoped* would work.
The Demand Gen campaigns were a complete disaster. The top-of-funnel campaign was bringing in users at nearly £18 a pop, and the retargeting campaign was even worse at over £23. For an app with a $14 entry price, this was just a non-starter. We were losing money on every single conversion. A lot of agencies would try to justify this spend under the banner of "brand awareness" or "building an audience," but that’s just jargon for wasting money. The numbers were rubbish, so we made the call to slash the budget on them almost immediately. We didn't get sentimental about it; we binned them.
On the other end of the spectrum, the winners were just as obvious. The App Install campaign was the standout star, pulling in new installs for an unbelievable 30p each. The Performance Max campaign was also doing a solid job, acquiring web registrations for £1.20. And the Search campaign, while the most expensive of the three winners at £2.60 per registration, was bringing in high-intent users who were actively looking for a solution like ScoutIQ. Each of these three campaigns was delivering users at a cost that made sense for the business.
So, the strategy became simple and ruthless: starve the losers and feed the winners. We took the budget we’d saved from the failing Demand Gen campaigns and reallocated it to the three that were performing. We pushed more ad spend into the App Installs, PMax, and Search campaigns to see how far we could scale them before the costs started to creep up. This process of aggressive budget reallocation is what separates a static, underperforming account from one that constantly improves. You can't be timid. When something is clearly working, you have to double down on it, hard and fast.
I've actually recorded a quick walkthrough of the account from around that time, showing exactly how we looked at the data in the dashboard and made these decisions. It gives you a real look under the bonnet at how we identified the winners and losers and shifted the budget around. You can watch it below.
Deep Dive: What Made the Search Campaigns Work?
While the App campaign drove volume, the Search campaign was our engine for acquiring high-quality web users. These are people who didn't just stumble upon an ad; they went to Google with a problem and typed in a query. They are much further down the funnel and more likely to convert into paying customers. Getting this channel right was a big part of the project, even if the cost per user was higher than the other channels.
Our strategy wasn't just to dump a list of keywords into an ad group. We structured the entire campaign based on user intent, mirroring the classic marketing funnel: Top of Funnel (TOFU), Middle of Funnel (MOFU), and Bottom of Funnel (BOFU). This allows you to tailor your messaging and bids based on how close someone is to making a decision.
Search Campaign Layering Framework
Layer 1: Awareness (TOFU)
Used broad match keywords like "sell books online". This cast a wide net to find people early in their journey who might not know an app solution exists. It's high risk, but a tight negative keyword list helps find new, untapped audiences.
Layer 2: Consideration (MOFU)
Used phrase match keywords for high-intent searches like "book scanner app" and "app for book value". These users know what they want; our job is to show them why ScoutIQ is the best option.
Layer 3: Decision (BOFU)
Targeted competitor brand names (like "ASellerTool") to capture users who are ready to buy but comparing options. Also included our own brand keywords to defend against competitors bidding on our name.
Our TOFU ad group, named `LH TOFU Selling Online`, was designed to catch people at the earliest stage. We used broad match keywords here, targeting general queries like "sell books online". Now, most advertisers are terrified of broad match, and for good reason. If you're not careful, Google will match your ads to all sorts of irrelevant rubbish, and you'll burn through your budget in hours. But, if you manage it properly with a very strict and constantly updated list of negative keywords, it can be an amazing tool for discovery. It lets Google's algorithm find new types of search queries that you would never have thought of. In our case, it was a huge success. This single ad group spent over £1,100 and brought in 433 conversions at a cost of just £2.55 each, making it the workhorse of the entire search campaign.
The MOFU ad group, `LH MOFU High Intent`, was more traditional. This is where we targeted people who knew what kind of tool they were looking for. We used phrase match for keywords like `"book selling app"` and `"scanner for books to sell on amazon"`. Phrase match gives you a good balance of control and reach, ensuring your ads show up for relevant searches without being too restrictive. This ad group brought in over 110 conversions at a similar cost of £2.77, proving its value in capturing that core, problem-aware audience.
Finally, our BOFU ad groups were all about the final decision. We had one targeting people searching for competitors like ASellerTool and Scoutly. The logic here is simple: if someone is looking up your competitor by name, they are a hot lead. You just need to get in front of them with a better offer. We also had an ad group for our own brand keywords, which is just good housekeeping. It ensures you dominate the results when someone searches for you directly and prevents competitors from stealing that traffic.
By splitting the campaign this way, we weren’t just throwing everything at the wall to see what stuck. We had a clear strategy for each level of user intent, which allowed us to speak to each user in the right way and, ultimately, bring in a steady stream of high-quality web registrations at a cost that worked.
The App Install Engine: Volume at a Ridiculous Price
Now let's talk about the campaign that brought in the lion's share of the users: the App Installs campaign. The final numbers were pretty staggering: 2,490 installs from a spend of just £754. That works out to a cost per install (CPI) of £0.30. It’s the kind of metric that makes marketing managers very happy.
But I have to be blunt here. Whenever you see a CPI that low from a Google Ads App campaign, you need to take it with a huge pinch of salt. The quality of these installs is always questionable. Google’s algorithm, when you set the goal to "installs," does exactly what you ask it to do: it finds the people who are most likely to click the "install" button, for the lowest possible price. It does not care one bit whether that person ever opens the app, signs up, or becomes a paying customer. You will get a lot of low-quality users who install the app and forget about it a minute later.
So why did we run it, and why did we scale it? Because for a business like ScoutIQ, the sheer volume can make up for the low quality. It's a numbers game. The app has a free trial and a relatively low entry price of $14. We knew that a huge percentage of these 30p installs would be useless. But we also knew that if even a tiny fraction of them opened the app, found it useful, and converted to a paid plan, the whole strategy would pay for itself many times over. It’s like a firehose approach to filling the top of your funnel. Most of the water will miss, but you'll still fill the bucket faster than with a dripping tap.
This strategy isn't for everyone. If you have a high-touch sales process or a very expensive product, a low-quality, high-volume approach like this would be a complete waste of time and money. But for an app with a self-service model, it can be an incredibly efficient way to get your product in front of a massive audience and let the best users filter themselves out. The blended model worked because these super-cheap installs pulled the overall average cost per user down to that headline-grabbing £0.96 figure, while the Search and PMax campaigns were working in the background to bring in the higher-quality users who were more likely to become long-term customers.
Monthly Cost Per Conversion Optimisation
This chart shows the impact of our continuous optimisation and budget reallocation. As we shifted spend to the winning campaigns and refined our targeting, the overall account efficiency improved dramatically month over month.
Putting It All Together: A Blended Acquisition Strategy
The real success of this project wasn't down to one single "killer" campaign. It was about how the different campaigns worked together as a system, each playing a specific role to contribute to the overall goal. Looking at any one campaign in isolation gives you a completely misleading picture. You have to look at the blended result.
Think of it like a football team. The Search campaign was the star striker, scoring high-quality goals (web registrations) from people who were already in a position to shoot. It was reliable but expensive. The App Install campaign was the midfield, just flooding the opponent's half with an overwhelming number of passes (installs). Many of them didn't lead to anything, but the sheer volume created constant pressure and opportunities. And the Performance Max campaign was the utility player, popping up all over the pitch, finding unexpected openings on YouTube, Gmail, and the Display Network that the other, more specialised players were missing.
Each part had a job to do:
- Search: Captured the highest-intent users for a higher, but acceptable, cost. This brought in the most qualified leads.
- App Installs: Drove massive volume and awareness at an extremely low cost. This brought down the overall blended CPA significantly.
- PMax: Acted as a discovery engine, efficiently finding new customers across Google's vast network.
When you blend the results of these three winning campaigns, you get the final picture. We acquired a total of 3,543 users for a total spend of £3,387. That’s how we arrived at the blended average cost per user of £0.96. The extremely cheap app installs offset the more expensive, but higher-quality, web registrations from Search. This portfolio approach gave us the best of both worlds: quality *and* quantity, at a price that was highly profitable for the business.
Total User Acquisition by Source
This breakdown shows how our multi-channel strategy contributed to the final user numbers. The high-volume App campaign was the primary driver, balanced by high-quality web registrations.
The Bottom Line: Why a Proper Strategy Beats Guesswork
What this all comes down to is that you can't just throw money at Google Ads and hope for the best. In 2026, the platform is too complex and too competitive for that. The days of setting up one simple campaign and watching the sales roll in are long gone. Winning requires a proper, structured process.
Our approach for ScoutIQ wasn't based on a secret "hack" or a magic button. It was based on a simple, repeatable methodology:
- Strategy First: We started with a plan, understanding the audience and mapping out our tests before spending a single pound.
- Test Broadly: We didn't assume we knew which campaign type would work best. We tested multiple channels across the Google network to let the data decide.
- Analyse Honestly: We looked at the numbers without emotion. The Demand Gen campaigns weren't working, so we cut them. The other three were, so we scaled them.
- Optimise Relentlessly: This isn't a one-time job. As the monthly performance chart shows, we were constantly refining things, which pushed the cost per conversion down by over 70% in just four months.
This is what separates campaigns that generate a real return from those that are just expensive hobbies. It’s about building an advertising *system*, not just running ads. It requires expertise and a disciplined, data-driven approach that most business owners simply don't have the time to develop.
If you're running Google Ads and you feel like you're stuck, or your costs are too high, or you just don't know where to go next, it's likely because you're missing a piece of this strategic puzzle. If you're interested in seeing what a similar, structured approach could look like for your business, we can definitely help. We offer a free strategy consultation where we can take a look at your account, crunch some numbers, and figure out the right strategy for you. It's a no-obligation chat to see if we can put a proper system in place to get you the results you're after.
Cite this Case Study
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.