What AI can (and can't) do for your app growth
Share on socials
What AI can (and can't) do for your app growth
Jump to Section
Jump to section
1. Metadata optimisation
2. Ad copywriting
3. Media buying
4. Copywriting
5. Review management
6. Campaign reporting
7. Product-market fit
8. Data analysis
Final thoughts
It feels like it's been a long time coming, but the sun is finally shining here in the UK!
So as I walked through the bustling streets of central London with coffee in hand, on what felt like the first true day of Spring, getting ready to attend the App Promotion Summit for the very first time, I wondered to myself, 'What else might be shining at this year's conference?'.
As I prepared to join hundreds of other marketers and app leaders for a day of talks, demos, and learning, I couldn't wait to see what was going to steal the spotlight.
No surprise, AI were the two letters on everyone's lips.
It was fascinating to learn where others, inside and outside of our industry, were on their AI journey. There was a real mix of eagerness to not only embrace but accelerate change, mixed with caution, and a warning that AI can't do everything for you. But what stood out most wasn't the hype. It was the honesty.
AI isn't magic; it's math. This was the quote of the day for me. The real magic happens when you combine AI, specifically generative AI, with expert teams who know how to ask the right questions, sense the story behind the data, and connect with real people.
Think of AI today as the world's most motivated intern—fast, tireless, and always learning. But it still needs a mentor. And if you're wondering what AI can and can't actually do when it comes to growing your app, here's a breakdown based on what I heard, saw, and jotted down at this year's Summit.
1. Metadata optimisation
Can: Help you do keyword research fast, with smart recommendations on search volume and competition. Tools like AppTweak and MobileAction are already using AI to improve visibility in store listings.
Can't: Choose the right keywords in context. AI can't assess what makes sense for your app's unique positioning, or the strategic trade-off between search volume and keyword relevance. That's still a job for you.
In fact, 65% of organisations have now adopted generative AI to support tasks like content creation and metadata updates, with marketing and sales seeing the highest adoption rates. However, the brands doing it best are guiding the AI with a clear understanding of their market and messaging.
2. Ad copywriting
Can: Draft multiple headlines and variations in seconds. It is brilliant for testing creativity quickly across platforms. Tools like Copy.ai or Jasper make it easier to keep your pipeline flowing.
Can't: Deliver the 'big idea' campaign that makes people feel something. AI lacks emotional intelligence and creative ambition. It can write a line, but not the line that gets people talking.
86% of marketers say AI makes them more efficient or saves time, but that doesn't mean it replaces the role of a creative director or brand strategist.
3. Media buying
Can: Handle the maths. AI can dynamically optimise bids, adjust budgets, and analyse campaign performance in real time at a speed and scale no human could match.
Can't: Make strategic trade-offs between short-term performance and long-term brand value. It doesn't understand your priorities, your seasonal plans, or where to flex the budget to align with your north star.
This came up repeatedly in London. Smart marketers are using AI for execution, but the decisions still need a human hand.
4. Copywriting
Can: Produce product descriptions, feature summaries, and onboarding text that's (generally) grammatically sound and quick to generate.
Can't: Nail your tone of voice without clear examples or prompts. AI doesn't get your brand personality unless you teach it, and it won't intuitively understand whether your audience prefers 'quick and simple' or 'in-depth and technical'.
Even the best AI writing tools need a style guide and review process because in a crowded app marketplace, tone is often what sets you apart.
5. Review management
Can: Automate replies in multiple languages, saving hours every week. If you're managing thousands of reviews a month, that time adds up fast.
Can't: Respond with empathy or humour. It won't know when a frustrated review needs a human follow-up, or how to use storytelling to turn a 3-star review into a 5-star one.
94% of companies believe AI will enhance personalisation, but real emotional resonance? That's still your edge.
6. Campaign reporting
Can: Automatically generate performance summaries and dashboards, helping teams move faster and get a pulse check at a glance.
Can't: Interpret results in the context of brand health, seasonality, competitor activity, or platform algorithm changes. AI will give you data, not insight.
AI can answer 'what happened?' but not always 'why it matters'.
7. Product-market fit
Can: Help you brainstorm new app ideas or identify emerging trends through social listening and market research tools.
Can't: Solve your product's usability issues. It won't know why users are dropping off after onboarding, or what's missing from your feature set. For that, you need user research and real feedback.
8. Data analysis
Can: Crunch huge data sets, spot anomalies, and surface patterns faster than any human could.
Can't: Tell you what to do with flawed or biased data. If the input is off, the output will be, too. Your team's experience is vital in asking the right questions and interpreting what the data means for your strategy.
Personalisation powered by AI can lift sales by up to 15%, but only when it's grounded in good data and guided by clear goals.
Final thoughts
AI is no longer a 'future trend'. It's already at work in many of the tools you use today, helping marketers improve Return on Ad Spend (ROAS), reduce Cost Per Acquisition (CPA), and get campaigns out the door faster.
But if there's one big takeaway from this year's App Promotion Summit, it's this: AI works best not when it replaces people, but when it works with people. The magic isn't in the model; it's in the team that knows how to use it.
AI isn't coming for your job, but it is changing it. Brands seeing the biggest gains from AI aren't just experimenting with tools. They're embedding AI into their workflows while keeping a clear handle on strategy, brand, and human judgment.
The truth is, AI excels at execution. It's brilliant at speeding up the 'doing' (writing variants, analysing data, generating reports, etc.). But it struggles with the 'deciding', figuring out what matters most, and why. That's where your expertise comes in.
So instead of asking, 'What can AI replace?', the better question is: 'Where can AI add value, and where do we need human intuition, creativity, and context to take it further?'.
The future of app growth isn't AI vs. humans. It's AI plus humans. AI as your fast-learning intern, and your team as the creative, strategic leads. Combine both, and you don't just move faster, you move smarter.
So, if you haven't already, now's the time to experiment, get comfortable with the tools, and work out where AI can support your ROAS, CPA, and growth goals. Just remember that the best results still come from teams that know their brand, understand their users, and aren't afraid to ask ‘why?'.
Written by
AI
Upscale