Hi there,
Thanks for reaching out!
Happy to give you some of my initial thoughts on getting app downloads with Google Ads. It can be a bit of a minefield if you're new to it, but the key is to forget about 'visibility' and focus entirely on getting the right people to install and use your app. A lot of people waste money chasing impressions when they should be chasing profitable actions.
I'll walk you through how I'd approach this, from the high-level strategy right down to the campaign setup.
TLDR;
- Stop thinking about 'visibility'. Your only goal should be driving high-quality app installs at a profitable cost. Optimising for anything else is a waste of money.
- You MUST use Google's dedicated 'App campaigns'. Don't try to use Search or Display campaigns; they aren't built for this and you'll just burn cash.
- Your ad creatives (images, videos, text) and your app store listing are more important than any targeting setting. The algorithm needs a wide variety of high-quality assets to find what works.
- The most important advice is to understand the maths behind your user acquisition. Use our LTV calculator below to figure out how much you can actually afford to pay for a new user.
- Success comes from feeding the algorithm the right data. That means setting up in-app conversion tracking properly from day one, so Google knows what a 'good' user looks like.
You're Paying Google to Find the Wrong People
Here’s the first bit of brutally honest advice: if you set up a campaign to get 'maximum visibility', you are actively paying Google to find the worst possible audience for your app. It sounds backwards, but it's true.
When you tell the algorithm your goal is 'Reach' or 'Impressions', you give it a simple command: "Find me the most eyeballs for the least amount of money." Google, being the efficient machine it is, does exactly that. It goes out and finds all the users who are browsing, scrolling, but are statistically very unlikely to ever click, install, or spend money. Why? Because their attention is cheap. No one else is bidding for them.
We saw this with a client years ago who insisted on running brand awareness campaigns. They got millions of impressions, the vanity metrics looked great, but their sales were flat. It was a classic case of paying a fortune to advertise to people who would never buy.
The only way to get downloads is to tell Google that downloads are what you want. You must use a campaign objective that optimises for conversions (in your case, installs or, even better, in-app actions). Forget visibility. Awareness is a byproduct of acquiring happy users who love your product, not a prerequisite for it. Every penny of your budget should be focused on getting someone to hit that install button.
There is Only One Campaign Type That Matters
For promoting an app on Google, there's really only one choice: Google App campaigns. Don't let anyone tell you to run a standard Search campaign pointing to your app store page, or a Display campaign with app store links. It's a complete waste of time and money. The platform has a purpose-built tool for the job, and you need to use it.
App campaigns are different from other types. You don't pick keywords or placements manually. Instead, you provide the building blocks:
- -> Text ideas: Headlines and descriptions about your app.
- -> Images: Screenshots, banners, lifestyle photos.
- -> Videos: Short clips showing the app in action.
- -> HTML5 assets: If you have them.
You then tell Google your goal (e.g., maximise installs), set a target cost per install (tCPI), and let the machine learning do the rest. It will automatically mix and match your assets and show them across Google's entire network – that includes Google Search, Google Play, YouTube, Discover, and the Google Display Network. It tests thousands of combinations to find what works best, on which network, for which user. Your job isn't to be a master tactician picking keywords; your job is to be a great creative director, feeding the machine with high-quality assets to test.
Step 1: The Goal
Define your primary objective. Is it just installs, or a specific in-app action like a subscription or level completion?
Step 2: The Assets
Provide a diverse mix of high-quality videos, images, and text. This is the fuel for the algorithm.
Step 3: The Bidding
Set a target Cost Per Install (tCPI) or Cost Per Action (tCPA). This tells Google how much you're willing to pay.
Step 4: Machine Learning
Google's AI tests asset combinations across all networks (Search, Play, YouTube, etc.) to find the most efficient path to your goal.
You Can't Out-Target a Bad Message
This is probably the most important part. People get obsessed with bidding strategies and campaign settings, but they neglect the one thing the user actually sees: the ad itself. A brilliant campaign structure won't save a boring ad. Your message and your offer are everything.
You need to speak directly to the user's problem. Don't sell "a new fitness tracking app"; sell "the motivation to finally stick to your workout plan". We use a simple framework for this called Before-After-Bridge.
- Before: Describe their current world. What's their pain point? "Struggling to find time for the gym? Feel like your fitness goals are always just out of reach?"
- After: Paint a picture of the world with your app. What's the dream outcome? "Imagine getting a personalised 20-minute workout you can do at home, seeing your progress every day, and feeling more energised than ever."
- Bridge: Position your app as the bridge to get them there. "Our app is the bridge. Get AI-powered workout plans that fit your schedule. Download now and start your free 7-day trial."
This structure works because it focuses on the user's transformation, not your app's features. Features are boring. Results are exciting.
For creative assets, variety is the most important thing. You need to supply the algorithm with lots of different options to test. I'd recommend having at least:
- -> 2-3 landscape videos (ideal for YouTube). Show the app in use, maybe a quick user testimonial.
- -> 2-3 portrait videos (for mobile placements). Keep them short and punchy, designed for sound-off viewing. UGC (User-Generated Content) style works wonders here. I remember one campaign for an events app where we saw the cost per signup drop significantly just by switching from slick, produced ads to simple iPhone videos of people using the app at a festival.
- -> 10-15 static images. A mix of clean screenshots, graphics highlighting a key benefit, and lifestyle images showing someone enjoying the benefit of your app.
The more you give the system to work with, the faster it can learn and the lower your costs will be.
We'll Need to Look at the Maths: What's a Download Worth?
Before you spend a single pound, you need to understand your numbers. The key question isn't "How cheap can I get a download?" but "How much can I afford to pay for a user and still be profitable?". This is where Lifetime Value (LTV) comes in.
For an app, LTV is the total profit you expect to make from a single user over their entire time using your app. To calculate it, you need a few key metrics:
- Average Revenue Per User (ARPU): How much money does the average user generate per month (from purchases, subscriptions, ad revenue, etc.)?
- Gross Margin %: What's your profit margin on that revenue?
- Monthly Churn Rate: What percentage of users do you lose each month?
Let's run an example. Say you have a subscription app:
- ARPU = £10/month
- Gross Margin = 70% (after app store fees etc.)
- Monthly Churn = 20% (meaning the average user sticks around for 1 / 0.20 = 5 months)
The calculation is: LTV = (ARPU * Gross Margin) / Churn Rate
So, LTV = (£10 * 0.70) / 0.20 = £7 / 0.20 = £35.
This means, on average, each new user is worth £35 in profit to you. A healthy business model usually aims for an LTV to Customer Acquisition Cost (CAC) ratio of at least 3:1. This means you can afford to spend up to £35 / 3 = ~£11.67 to acquire a new user. This number is your North Star. It tells you what your target Cost Per Install (CPI) should be. Suddenly, a £2 CPI doesn't just look good, it looks incredibly profitable.
I've built a simple calculator for you to play with these numbers. It'll help you get a feel for how small changes in churn or revenue can dramatically impact how much you can afford to spend on ads.
User Lifetime Value (LTV)
£35.00Max Affordable CPA (3:1 Ratio)
£11.67I'd say you should structure your account like this...
Now for the practical bit. Setting up the campaign is fairly straightforward if you follow a logical process. The absolute first thing you must do is ensure your conversion tracking is flawless. You need to be tracking not just installs, but also the key actions within your app that signal a valuable user (e.g., tutorial completion, subscription start, first purchase). This is the data you feed back to Google to teach it what you want more of.
Once tracking is solid, here's the plan I'd lay out:
- Start with one App Campaign. Don't overcomplicate it. One campaign to begin with.
- Set the Bidding Strategy. Start with "Target cost per install" (tCPI). Use the "Max Affordable CPA" from the calculator above as your starting point. You're telling Google, "I don't want to pay more than £X for a new user". Once you have a good volume of in-app conversions (say, 50+ per week of your main goal, like a subscription), you can think about switching to "Target cost per action" (tCPA) to optimise for higher-quality users, not just installs.
- Location Targeting. Start with your most valuable country. If you're based in the UK and most of your potential customers are here, start with just the UK. Don't target the whole world from day one; you'll get a lot of low-quality traffic and it will pollute your data. A focused, strategic approach is what gets results. For instance, in one app growth campaign we managed, we generated over 45,000 signups while keeping the cost under £2 per signup.
- Asset Groups. This is where you upload your creatives. I would create at least two asset groups to start.
- Asset Group 1: General Messaging. Focus on the core value proposition of your app. Your best "Before-After-Bridge" copy goes in here, along with your strongest videos and images.
- Asset Group 2: Niche Angle. Focus on a different pain point or benefit. For the fitness app, maybe this angle is about "mental wellness" or "beating gym anxiety" rather than just physical fitness. This allows you to test different marketing angles against each other.
- Budget. Start with a budget you're comfortable with, but it needs to be enough for the algorithm to learn. I'd recommend a daily budget of at least 10x your target CPI. So if your tCPI is £5, you need at least £50/day. Anything less and the campaign will struggle to get out of the 'learning phase'.
After you launch, do not touch it for at least 7-10 days. The algorithm needs time to learn. Constantly tweaking things early on is the fastest way to kill a campaign's performance. After a week or so, you can start looking at the asset report. Google will tell you which headlines, videos, and images are performing 'Best', 'Good', or 'Low'. Your job then becomes a cycle of removing the 'Low' performing assets and replacing them with new ideas to test.
I've detailed my main recommendations for you below:
| Phase | Actionable Step | Why It's Important |
|---|---|---|
| 1. Foundation | Set up robust in-app conversion tracking for installs AND key post-install events (e.g., signup, trial start, purchase). | This is critical. Without this data, Google's algorithm is flying blind and cannot optimise for high-quality users. |
| 2. Strategy | Calculate your user LTV and determine your maximum affordable Cost Per Install (CPI). | This gives you a data-driven target for your bidding strategy and ensures your campaigns are built for profitability from the start. |
| 3. Campaign Setup | Create ONE Google App campaign targeting your primary country. Start with a tCPI bidding strategy based on your LTV calculation. | Keeps things simple and allows the algorithm to focus its learning on your most important market first. |
| 4. Creative | Upload a wide variety of assets (at least 5 videos, 10 images, 5 headlines/descriptions) based on the 'Before-After-Bridge' framework. | The algorithm's performance is entirely dependent on the quality and variety of the creative assets you provide. More assets = more tests = faster learning. |
| 5. Optimisation | Let the campaign run for 7-10 days untouched. After that, begin a weekly process of swapping out 'Low' performing assets with new creative ideas. | Avoids premature changes during the crucial learning phase. Focuses on the single most impactful lever for improvement: creative testing. |
This whole process isn't just about setting up an ad and hoping for the best. It's a systematic approach to understanding your audience, crafting a compelling message, and using the platform's technology to your advantage. It takes time and a bit of budget to get right, but it's a far more reliable path to growth than just boosting posts or running 'visibility' campaigns.
Getting this right can be the difference between burning through cash and building a sustainable user acquisition engine. If you'd like an expert pair of eyes on your specific app and strategy, we offer a free initial consultation where we can look at your situation in more detail. It's often helpful to have someone who's done this many times before to point out the quick wins and potential pitfalls.
Hope this helps!
Regards,
Team @ Lukas Holschuh
