Hi there,
Thanks for reaching out!
I’ve had a look at your question about Google App Ads creative for the Cambridge market. It’s a common stumbling block, this idea that a specific city has some magic code that needs cracking. I’m happy to give you some initial thoughts and guidance on this. The truth is, your problem probably has very little to do with Cambridge itself and a lot more to do with your fundamental approach to strategy, creative, and data. Let's unpack that a bit.
TLDR;
- Stop blaming Cambridge. Your problem isn't the "local nuance" of a specific city; it's almost certainly a generic strategy that isn't resonating with anyone, anywhere.
- Your Ideal Customer Profile (ICP) is not a demographic; it's a specific, urgent problem state. You need to define the nightmare you solve for a Cambridge student, tech worker, or tourist.
- Creative performance is a direct result of your campaign structure and data inputs. You're likely not giving Google's algorithm enough varied assets or the right conversion goals to work with.
- The most important advice is to stop guessing and start testing with a proper framework. Build creative for specific personas (we'll build some examples below) and let the data tell you what "local" actually means.
- This letter includes an interactive ROAS calculator to help you understand the real maths behind a profitable app campaign.
We'll need to look at your strategy, not your geography...
Let's be brutally honest. Focusing on "Cambridge nuances" is a distraction. It's what people do when the core message isn't working. You could be advertising in Cambridge, Coventry, or Cardiff – if your underlying strategy is weak, the results will be the same: wasted money. The laws of good advertising don't change at the city limits.
The real issue is almost always a failure to define the customer by their *pain*. Your Head of Engineering client at the Science Park isn't just a job title; she's a leader terrified of her best developers quitting out of frustration with a broken workflow. Your student user at Anglia Ruskin isn't just a 'student'; they're someone stressed about finding an affordable meal during exam week so they can focus on their dissertation. Your ICP isn't a person; it's a problem state.
Before you spend another pound on ads, you have to answer this: What specific, urgent, expensive, career-threatening, or life-complicating nightmare does your app solve? Forget demographics like "people aged 20-30 in Cambridge." That tells you nothing. Instead, think in terms of problems:
- Nightmare A: "I'm a PhD student at Trinity College, I'm completely swamped, and I can't find a reliable, quiet study space outside of the library that's open late."
- Nightmare B: "I've just moved to Cambridge to work for AstraZeneca, I don't know anyone, and I'm finding it impossible to meet people outside of work."
- Nightmare C: "We're visiting Cambridge for the weekend, and every restaurant we find online is either a tourist trap or fully booked. We just want an authentic, good meal."
See the difference? These are real problems felt by real people. Your creative doesn't need to be "Cambridge-y." It needs to be a direct answer to one of these nightmares. Once you've isolated that pain, your entire targeting and messaging strategy becomes clear. You're not targeting "Cambridge"; you're targeting the manifestation of a universal human problem within a specific geographic area. It's a subtle but absolutly critical distinction.
I'd say you need to fix your campaign structure first...
Before we even get to the colour of your buttons, we need to talk about how you're feeding the machine. Google App Ads is a black box powered by machine learning. Its performance is only as good as the ingredients you give it. If you give it one video, two headlines, and one image, you're asking it to cook a gourmet meal with a single carrot and a pinch of salt. It just won't work.
You need to stop thinking in terms of "making ads" and start thinking in terms of "supplying assets." The algorithm's job is to mix and match these assets to find the winning combinations for different sub-segments of your audience. Your job is to provide a rich, diverse pantry of ingredients. You should aim to max out the asset slots Google gives you:
- Headlines: Up to 5. These need to be varied. One should state the problem, one the benefit, one a key feature, one a call to action, and one should be social proof.
- Descriptions: Up to 5. These are longer-form versions of your headlines. Elaborate on the pain points and solutions.
- Images: Up to 20. This is crucial. Don't just upload 20 variations of your logo. You need screenshots of the app in action, lifestyle photos of your target user (e.g., a student on a laptop in a Cambridge cafe), UI mockups showing key features, and simple graphics with text overlays.
- HTML5 Ads: Up to 20. If you have the resource, these can be powerful.
- Videos: Up to 20. This is where you can make a huge impact. You need variety:
- UGC-style: Someone talking to their phone camera about how the app solved their problem. Raw and authentic works best.
- Screen recording: A simple walkthrough of the app's core function. No fancy production needed.
- Simple animation: A kinetic text video highlighting the main benefit.
- Landscape, Portrait, Square: You need assets in all the main formats to ensure your ad can run across all of Google's inventory (YouTube, Discover, Search, Display).
Furthermore, what is your campaign goal? Most people start with "Install Volume," which is a mistake. You are telling Google, "Find me the cheapest possible installs." The algorithm will dutifully find you users who love downloading apps and never using them. Their attention is cheap for a reason.
You MUST set up in-app conversion events (e.g., 'registration_complete', 'first_purchase', 'level_5_achieved') and set your campaign goal to "In-app actions." This tells Google, "Find me users who are likely to actually DO this valuable thing inside my app." The Cost Per Install (CPI) will be higher, but the quality of the user will be infinitely better. It's the difference between buying a list of random phone numbers and a list of qualified leads. For one of our clients, a medical job matching SaaS, we saw a reduction in their Cost Per User Acquisition from £100 down to £7. Focusing on user quality is that significant.
You probably should focus on the maths, not the myths...
Let's talk numbers. The success of an app campaign isn't measured in downloads; it's measured in profit. To understand this, you need to know your Customer Lifetime Value (LTV). If you don't know this number, you're flying blind and you have no business spending money on ads. The real question isn't "How low can my CPI go?" but "How high a CPI can I afford to acquire a great customer?"
Here’s the simple maths:
LTV = (Average Revenue Per User * Gross Margin %) / Monthly Churn Rate
Let's say your app has a subscription, and your Average Revenue Per Account (ARPA) is £10/month. Your margin is 90% (typical for software), and you lose 8% of your users each month (8% churn).
LTV = (£10 * 0.90) / 0.08 = £112.50
Each customer is worth £112.50 to your business. A healthy LTV:CAC (Customer Acquisition Cost) ratio is 3:1. This means you can afford to spend up to £37.50 to acquire a single paying customer. If 1 in 10 people who install and register eventually become a paying customer, you can afford to pay up to £3.75 per registration. Suddenly, that £2.50 Cost Per Action from the chart above doesn't look expensive at all. It looks like a bargain.
This is the maths that unlocks aggressive, intelligent growth. It frees you from the tyranny of chasing cheap, low-quality installs and allows you to focus on acquiring users who will actually contribute to your bottom line. Use the calculator below to get a feel for your own numbers. It might be an eye-opener.
App Campaign Profitability Calculator
You'll need creative that speaks to a problem, not a place...
Right, now that we've sorted the strategy, structure, and maths, we can finally talk about creative. But we're not going to make "Cambridge ads." We're going to make ads for specific people with specific problems who happen to live in Cambridge.
Let's build out three distinct personas and brainstorm some actual creative concepts. Notice how none of these rely on generic shots of King's College Chapel.
Persona 1: The Overwhelmed Post-Grad Student
- Who: Sofia, 25, doing her MPhil at Darwin College.
- Nightmare: She's drowning in reading lists and dissertation prep. Her biggest friction is wasted time on daily life admin, especially figuring out what to eat. She's budget-conscious but values her time more.
- App Solution (Hypothetical): An app that delivers healthy, affordable, pre-prepped meals specifically for students.
Creative Concepts (Problem-Agitate-Solve):
Headline 1: Dissertation deadline looming?
Headline 2: Stop wasting time on cooking.
Headline 3: Healthy meals for students, delivered.
Description: Buried in books at the library and living off toast? Get your study time back. We deliver brain-fuel meals designed for students, right to your college door. Focus on your finals, not your frying pan.
Video Idea: A vertical, UGC-style video. Shot on a phone. Starts with a close-up of Sofia looking stressed, surrounded by books. She says, "I've got three essays due and a dissertation proposal, and honestly, I haven't eaten a proper vegetable in a week." Quick cut to her unboxing a healthy-looking meal from the app. She takes a bite. "This is a lifesaver. Takes two minutes and I can get straight back to work." Simple, relatable, effective. No mention of Cambridge, but it speaks directly to thousands of people there.
Persona 2: The Recently Relocated Tech Professional
- Who: Ben, 32, a software engineer who just moved from London to work at ARM.
- Nightmare: He's earning good money but his social life is non-existent. He finds Cambridge a bit cliquey and hard to break into. He's looking for ways to meet like-minded people with similar interests outside of work.
- App Solution (Hypothetical): A social app that connects people based on niche hobbies (e.g., board games, hiking, coding side-projects).
Creative Concepts (Before-After-Bridge):
Headline 1: New to Cambridge? It's tough.
Headline 2: Find your people. Not just colleagues.
Headline 3: From lonely nights to game nights.
Description: Before: Your Cambridge weekend is another night of Netflix. After: You're at a pub quiz with your new board game group. Our app is the bridge. Connect with people who share your actual interests, from bouldering to Python.
Image Idea: A split-screen image. On the left, a shot of a single person looking bored on their sofa, blue light from a screen on their face. On the right, a vibrant shot of a small group of people laughing together over a board game in a pub (like The Eagle). The visual contrast tells the whole story.
Persona 3: The Discerning Tourist
- Who: A couple, Mark and Sarah, 45, visiting from Manchester for a weekend.
- Nightmare: They want an authentic experience but feel like they are just being funnelled from one tourist trap to the next. They hate crowded chain restaurants and souvenir shops.
- App Solution (Hypothetical): A curated city guide app that focuses on independent businesses and hidden gems recommended by locals.
Creative Concepts (Attack the Pain):
Headline 1: Escape the tourist traps.
Headline 2: See Cambridge like a local.
Headline 3: Ditch the crowds. Find the gems.
Description: Tired of seeing the same 5 places on every travel blog? We get it. Discover the independent coffee shop down a cobbled alley, the pub the students actually drink in, and the best view of the river that isn't on a postcard. Your authentic Cambridge weekend starts here.
Video Idea: A fast-paced montage (30 seconds, vertical format for YouTube Shorts/Reels). Quick cuts between an overcrowded, generic tourist spot (e.g., a massive queue for punting) and then a beautiful, quiet, and appealing 'hidden' spot found via the app. Text overlays: "Not This... This." It directly addresses the visitor's desire for a unique experience.
This is the main advice I have for you:
To pull all of this together, here’s a more structured look at the actionable steps you should be taking. This isn't a checklist to be completed once; it's a continuous cycle of improvement. This is how you move from guessing about "nuances" to building a predictable, scalable customer acquisition machine.
| Problem Area | My Recommendation | Why It Matters | First Step to Take |
|---|---|---|---|
| Strategy | Stop focusing on geography. Define 2-3 core customer personas based on their specific, urgent "nightmare" or pain point. | A pain-point focus makes your ads universally relevant and powerful. It’s the foundation for all effective messaging and creative. | Sit down and write out the three "nightmares" your app solves, as we did in the examples above. Be as specific as possible. |
| Campaign Goal | Immediately switch your campaign objective from "Install Volume" to "In-app actions." Ensure your key in-app events are set up as conversions. | This tells Google to find you valuable users who will actually engage and monetise, not just cheap downloaders. It's the single biggest lever for improving user quality. | Go into your Google Ads account and your analytics platform (e.g., Firebase) today. Identify the most important user action (e.g., complete registration) and import it as a primary conversion goal. |
| Creative Assets | Adopt an "asset-first" mindset. Create a large and diverse library of headlines, descriptions, images, and videos in all formats (landscape, portrait, square). | Google's AI needs a wide variety of ingredients to find the best-performing combinations. A lack of assets is the most common reason for campaign underpeformance. | Create a simple spreadsheet. Rows for Headlines, Descriptions, etc. Columns for your different personas. Your goal is to fill this matrix with at least 50+ total assets. |
| Measurement | Calculate your estimated LTV and determine your target Cost Per Action (CPA) based on a 3:1 LTV:CAC ratio. Focus on this metric, not just CPI. | This gives you a clear, data-driven definition of success. You'll know exactly how much you can afford to pay for a quality user, enabling you to scale confidently. | Use the calculator in this letter with your best estimates for ARPU and churn. This will give you your North Star metric for campaign optimisation. |
| Iteration | Regularly review your asset performance report in Google Ads. Double down on winning themes/angles (what problems are resonating?) and replace the worst-performing assets every 2-4 weeks. | Creative fatigues quickly. Continuous testing and iteration is the only way to maintain and improve peformance over time. It’s a process, not a one-time setup. | Set a recurring calendar reminder for every two weeks to "Review & Refresh App Ad Assets." Treat it as a non-negotiable task. |
As you can see, this is a much deeper process than just trying to figure out what pictures will appeal to people in Cambridge. It's about building a robust, data-driven system. This kind of work is what separates the campaigns that fail after a few weeks from the ones that become scalable engines for growth.
Getting this right takes time, expertise, and a rigorous testing methodology. It's not just about setting up an ad and hoping for the best. It's about understanding your audience's deepest needs, optimising your targeting, creating a pipeline of compelling creative, and constantly analysing the data to refine your approach.
This is where professional help can make a huge difference. With years of experience running app campaigns, we can help you bypass the costly trial-and-error phase and implement a proven strategy from day one. We can provide insights you might not have thought of and take over the entire implementation and optimisation process for you, ensuring that every pound you spend is working to grow your user base profitably.
If you’d like to have a chat about how we could apply this kind of thinking specifically to your app, we offer a completely free, no-obligation 20-minute strategy session where we can review your current setup and provide some more tailored advice.
Hope this helps!
Regards,
Team @ Lukas Holschuh