Hi there,
Thanks for reaching out!
I've had a look over the situation with your Google App campaigns. It's a really common spot to be in, hitting a wall where things just feel 'stale' and you're not sure which lever to pull to get things moving again. Happy to give you some initial thoughts and guidance based on what we've seen work for other apps. The short answer is you're probably focusing a bit too much on tweaking the small things inside one campaign, when the real growth will come from thinking bigger about your overall strategy, creative, and channel mix.
Let's get into it.
TLDR;
- Your tROAS is a signal to Google. Lowering it when you're already underperforming is telling the algorithm to find you cheaper, lower-quality users, which will almost certainly drop your actual ROAS further.
- Stop trusting Google's estimations. They are notoriously unreliable and are designed to encourage more spending. Real-world testing is the only truth.
- Repeating assets with minor tweaks is inefficient and leads to creative fatigue. You need to test fundamentally different creative concepts and angles, not just variations of the same idea.
- The most important piece of advice is that you've likely hit a plateau on Google Ads. The key to breaking through is to expand to other high-intent channels like Meta Ads and Apple Search Ads to tap into new audiences.
- This letter includes an interactive calculator to help you work out your true Customer Lifetime Value (LTV), which is the only number you should be using to decide your ROAS targets.
We need to talk about that tROAS target...
Okay, first things first. Your main question about tROAS. You've set a target of 300% but you're only hitting 198%. Now you're wondering if lowering the target will get you more volume. The simple, brutally honest answer is: yes, it probably will get you more volume, but it'll be worse volume, and your actual ROAS will almost definately drop even further.
Think of it like this: your tROAS setting is a direct instruction to Google's algorithm. You're telling it, "For every £1 I give you, I need you to find me users who will generate at least £3 in return." Right now, the algorithm is struggling to do that. It's only finding users that return £1.98. If you then go and lower your target to, say, 200%, you're effectively telling it, "Okay, you were right, I was too ambitious. Forget finding me those high-value users, just go find me anyone who will generate £2."
The algorithm will breathe a sigh of relief and go after the lower-hanging fruit. The cheaper, less engaged users who are easier to convert but who won't spend as much in your app over time. So you'll see more conversions, sure, but your conversion value will drop, and your actual ROAS will likely settle somewhere *below* your new, lower target. It's a race to the bottom. I've seen it happen time and time again. A client comes to us with a ROAS problem, and the first thing we often see is that they've been chasing volume by continually lowering their targets, effectively training the algorithm to bring them rubbish.
So, how do you decide on a tROAS? You don't let Google's estimations guide you. You work it out from your own business metrics. You need to know what your profit margin is and what a new user is actually worth to you over their lifetime. Anything else is just guessing. We'll get into that a bit later with a calculator that should make it clearer.
You probably should stop trusting Google's estimations
This brings me to your next point. You mentioned you use Google's estimations a lot to decide on your tROAS. You need to stop. Right now. Those little predictions and simulators are a sales tool for Google, not a strategic tool for you. They are designed to do one thing: encourage you to spend more money by showing you optimistic, best-case scenarios.
The problem is, these estimations are based on historical auction data and broad modelling. They have absolutely no idea about:
- Creative Fatigue: They can't predict when your current ads will stop being effective.
- Competitor Actions: They don't know if a major competitor is about to launch a huge campaign in your target market and drive up costs.
- Market Seasonality: They can't account for real-world events or shifts in user behaviour.
- Your Specific Offer: Their models are generic; they don't truly understand the nuance of your app and why a user might choose you over someone else.
I've lost count of the number of accounts I've audited where the performance charts look nothing like the rosy picture Google's simulator painted. Tbh you can't really trust any of these predicions as they are often wrong. The only way to know what will happen if you change your bid is to actually change it and collect your own data. Test, measure, learn. That's the only ground truth in this game. Relying on Google's estimations is like asking the casino owner for tips on how to win at roulette. Their advice will always, ultimately, benefit the house.
I'd say you should fix your asset groups first...
Your question about repeating assets is a good one, and it hints at a bigger issue. Having many ad groups with slightly different tones for the same assets is generally not a good idea. You're creating a situation where your own ad groups are competing against each other for the same audience, and you're not giving the algorithm clear signals about what actually works.
Google App Campaigns work best when you give them a diverse range of distinct assets and let the machine learning figure out the best combinations. When you supply very similar assets, you're not really testing anything. You're just muddying the waters. The algorithm might report that 'Headline A' is slightly better than 'Headline B', but if they both communicate the exact same message, you haven't learned anything meaningful about your audience.
A much better approach is to structure your tests around completely different *concepts* or *angles*. Instead of testing "The best app for productivity" vs. "The greatest productivity app," you should be testing concepts like:
- Angle 1: Time-Saving. Focus all your assets (images, videos, headlines) on how your app saves the user time. (e.g., "Get an hour back every day," "Finish your to-do list by lunch")
- Angle 2: Collaboration. Focus everything on how the app helps teams work better together. (e.g., "The seamless way to manage team projects," "Never miss a team deadline again")
- Angle 3: Feature-Specific. Pick your one killer feature and build all assets around showcasing it. (e.g., "Our AI scheduler plans your week for you," "Instant file syncing across all devices")
You should create one ad group (or asset group in this case) per concept. This gives Google a clear, distinct set of assets to test. After a week or two, you'll be able to see which fundamental message resonates with your audience, not just which synonym performs 2% better. Then you can double down on the winning angle and start testing new ones. This methodical approach is how you get real, actionable learnings that actually improve performance, rather than just spinning your wheels with minor tweaks.
This is how I would structure it:
| Campaign Element | Asset Group 1: Time Saving Angle | Asset Group 2: Collaboration Angle | Asset Group 3: AI Feature Angle |
|---|---|---|---|
| Core Message | "Save an hour every day." | "Effortless team project management." | "Let our AI plan your perfect week." |
| Headlines | - Reclaim Your Schedule - The Fastest To-Do List - More Done, Less Stress |
- Sync With Your Team - Project Tracking, Simplified - Better Teamwork Starts Here |
- Smart AI Assistant - Auto-Schedule Tasks - The Future of Planning |
| Video Assets | - Screen recording showing a task being completed in seconds - User testimonial about time saved |
- Animation of a team collaborating in-app - Split-screen showing seamless updates |
- Demo of the AI feature in action - "Before & After" of a messy vs organised schedule |
| Image Assets | - A clock with a checkmark - A person relaxing, their work done |
- Graphic showing interconnected user profiles - A clean, shared dashboard |
- A robot icon or brain graphic - A sleek UI screenshot of the AI feature |
You'll need a way to break this plateau...
The fact that your results are 'stale' is the biggest clue here. It tells me you've likely saturated your current audience on Google Ads. You can tweak bids and assets all day long, but you're probably just shuffling the same users around. When you hit this point, the answer isn't to squeeze the lemon harder; it's to find a new lemon.
You need to expand your channel mix. Google App Campaigns are great for capturing intent, but they are just one piece of the puzzle. We've seen massive growth for apps by layering on other platforms. For instance, one campaign we ran brought in over 45,000 signups at under £2 per signup by using a combination of Google Ads, Meta Ads, TikTok Ads, and Apple Ads. Another software client saw 5,082 trial signups at just $7 each using Meta Ads.
Here's how you should be thinking about it:
- Meta Ads (Facebook & Instagram): This is your next logical step. The targeting capabilities are immense. You can build lookalike audiences from your best existing users, target people based on interests related to your app's category, and retarget people who visited your website or App Store page but didn't install. It's brilliant for creating demand, not just capturing it. The cost per install can often be lower than Google, especially when you find a winning creative.
- Apple Search Ads (ASA): If you're on iOS, this is a non-negotiable. It's pure bottom-of-the-funnel intent. You're placing your app at the very top of the App Store search results when someone is actively looking for an app like yours. The quality of users from ASA is often the highest of any channel because they are in the process of downloading *right now*. It's a must-have for any serious app marketing strategy.
- TikTok Ads: Don't dismiss it. Depending on your app and audience, it can be a huge driver of volume at a very efficient cost. It requires a different style of creative – more native, authentic, and user-generated – but the potential reach is enormous.
Sticking only to Google is like only fishing in one corner of a giant lake. There are plenty more fish out there, you just need to use different bait and cast your line in a new spot. Start by allocating 15-20% of your current budget to testing Meta Ads and see what happens. I suspect you'll be suprised by the results.
You'll need to know your numbers inside out...
Let's circle back to the most important question: how to decide your tROAS. As I said, it has to be based on your actual business economics. The key metric you need to understand is your Customer Lifetime Value (LTV). This tells you how much gross profit a typical user generates for you over their entire time using your app.
The real question isn't "How low can my Cost Per Install go?" but "How high a CPI can I afford to acquire a truly great user?" Once you know your LTV, you can answer this confidently. Let's break it down:
- Average Revenue Per User (ARPU): What do you make per user, per month? (From subscriptions, in-app purchases, etc.)
- Gross Margin %: What's your profit margin on that revenue after app store fees and other direct costs?
- Monthly Churn Rate %: What percentage of users do you lose each month?
The basic formula is: LTV = (Monthly ARPU * Gross Margin %) / Monthly Churn Rate
For example, if your ARPU is £10, your margin is 70% (after Apple/Google's cut), and your churn is 20%, your LTV is (£10 * 0.70) / 0.20 = £35. This means you can spend up to £35 to acquire a user and still break even over their lifetime. A healthy target is to spend about a third of your LTV on acquisition, which gives you an LTV:CAC (Customer Acquisition Cost) ratio of 3:1. In this case, your target CAC would be around £11.67.
Your ROAS target is then derived from this. If an average user spends £5 in their first month, to get a CAC of £11.67, you'd need a ROAS of (£5 / £11.67) = 0.43x... which doesn't seem right for ROAS. Let's re-think. ROAS is Return on Ad Spend. So if you spend £11.67 and they generate £35 in their lifetime, your lifetime ROAS is 3x. But tROAS bidding is based on a shorter conversion window. If you're optimising for revenue within 7 days, and a new user generates on average £2 in that window, then to hit your target CAC of £11.67, you need a 7-day ROAS of (£2 / £11.67) = 17%. But that's not how tROAS works in the interface. A 300% tROAS means you want £3 back for every £1 spent. This means your Cost per Purchase is 1/3 of your Average Order Value.
Let's simplify. If your LTV is £35, and you want a 3x return, your target acquisition cost is ~£11.67. If the initial conversion you're tracking (e.g., first purchase) is worth £10 on average, then to acheive a CAC of £11.67, your ROAS needs to be (£10 / £11.67) = 85%. Setting a tROAS of 300% (or 3x) implies you're only willing to pay £3.33 for that £10 initial purchase. This might be too restrictive and is likely why the algorithm is struggling. You might be strangling your campaigns by being too conservative with your tROAS, based on short-term value rather than lifetime value.
Here’s an interactive calculator to help you figure out your own LTV. Play with the sliders to see how small changes in churn or revenue can massively impact what you can afford to spend on ads.
This is the main advice I have for you:
To sum up, you're at a critical point where small tweaks won't cut it anymore. You need a more structured, strategic approach to break out of this slump. Sticking with what you're doing will just lead to more stale results and wasted ad spend. It's time to test new channels, get serious about creative, and ground your bidding strategy in solid business maths, not platform estimations.
I've detailed my main recommendations for you below in a more actionable format:
| Area of Focus | Problem | Recommended Action | Why It Matters |
|---|---|---|---|
| 1. Bidding Strategy | tROAS is set too high for current performance, and lowering it is risky. Decisions are based on unreliable estimations. | Do not lower your tROAS yet. First, calculate your true LTV and affordable CAC. Adjust your tROAS based on that reality, potentially setting it lower but with a clear understanding of your profit margin. | This stops the "race to the bottom" and ensures your ad spend is profitable on a lifetime basis, not just in a 7-day window. It gives the algorithm a realistic goal. |
| 2. Creative Testing | Repeating similar assets across multiple ad groups is causing internal competition and providing no real learning. | Pause redundant asset groups. Restructure your campaigns to test 2-3 distinct creative 'angles' (e.g., pain point vs. benefit vs. feature). Dedicate one asset group to each angle. | This gives you clear data on what message actually resonates with your audience, leading to breakthrough performance instead of incremental (and often random) gains. |
| 3. Channel Expansion | Performance is "stale," indicating you've likely saturated your core audience on Google Ads alone. | Allocate 20% of your budget to launch new campaigns on Meta Ads and Apple Search Ads. Start with lookalike audiences on Meta and keyword campaigns on ASA. | This is the single biggest lever for growth. It opens up entirely new pools of users, reduces your reliance on one platform, and allows you to find customers at different stages of awareness. |
| 4. Settings & Oversight | Auto-apply settings are a known risk that can derail budgets and strategy without warning. | You've already done this, which is great. Continue to regularly audit your account for any new auto-apply "recommendations" that Google tries to enable by default. Take back full control. | Ensures your strategy is being executed as planned and prevents the algorithm from making budget-draining changes that aren't aligned with your goals. |
Implementing all of this correctly—from the maths to the multi-channel setup and creative strategy—is a significant amount of work, especially when you're also trying to develop your app. It involves a lot of testing, analysis, and experience to know which levers to pull and when. While you can certainly do this yourself over time, it's often faster and more effective to work with someone who has navigated these plateaus hundreds of times before.
We specialise in scaling app campaigns exactly like yours. If you'd like to have a more in-depth chat, we offer a free, no-obligation strategy session where we can go through your account together and map out a more concrete plan of action. It could be a really efficient way to get some clarity and get your growth back on track.
Hope this helps give you a new perspective!
Regards,
Team @ Lukas Holschuh
