An AI app growth marketer finds, tests and scales the channels that bring users into an AI app at a cost the business can afford. The job is a weekly loop: research, creative, launch, measure, kill or scale. I have done it since 2020, most recently taking ZuAI from 10,000 to 2 million users.
The definition
An AI app growth marketer owns user acquisition for an app whose core feature is an AI model. One person is responsible for the number of new users, what each one cost, and whether they stayed. The work covers paid ads, organic channels such as TikTok and Reddit, creative production, tracking, and the weekly call on what to kill and what to scale. It is an operator role, not an advisory one. If the number does not move, the marketer explains why.
Not the same as a generic growth marketer
A generic growth marketer might work on SaaS, a marketplace, or an ecommerce store. Sales cycles run weeks. Retention is measured in months. The feedback loops are slow.
AI apps are different in ways that change the daily work:
- Users decide in the first session whether the app is useful. There is no onboarding call to save them.
- Churn is fast. In my campaigns a user who does not come back in the first week rarely comes back at all.
- Willingness to pay is low. Most users expect AI to be free and treat a paywall as a reason to leave.
- Organic upside is huge. One 15 second video of the app doing something useful can beat a month of paid spend.
So the role spends more time on creative than on funnels, and more on activation than on lead nurture. The generic playbook is aimed at a different product.
Not the same as an agency
An agency runs your ads. An AI app growth marketer runs your growth. The difference shows up in three places.
- Scope. Agencies are paid per channel. They run TikTok or they run Meta. A growth marketer owns blended CAC across every channel, paid and organic, and is judged on that one number.
- Speed. Agencies work in monthly reports. A growth marketer works in weekly kills. A creative that dies on Tuesday is replaced on Wednesday, not discussed at month end.
- Incentive. An agency earns more when you spend more. A growth marketer earns the same at $3,000 a month or $300,000. There is no reason to keep a losing channel alive.
None of this is a knock on agencies. The usual problem is that nobody owned the whole number.
The weekly loop
Here is the job week to week. Every week, no exceptions.
Research. Read the app’s reviews, the Reddit threads where the problem comes up, and the comments on last week’s videos. Write down the exact words users use. Those are next week’s hooks.
Creative. Brief creators or shoot it yourself. New hooks, new formats, new scripts. On ZuAI this is 150 or more creatives a month. For a smaller app, 20. What matters is that it is new every week.
Launch. Put the creatives live across paid and organic. Fixed budget per creative. Same launch day each week so the data lines up.
Measure. Once a creative hits its spend threshold, read the numbers. Cost per install, activation rate, day 7 retention, blended CAC for the account.
Kill or scale. Anything under the line is turned off. Anything over it gets more budget and a variation for next week. Write one sentence on why the winner won.
Then it starts again. Not exciting. A year of it is.
Tools and channels
The channels I run for AI apps, in the order I usually test them:
- TikTok UGC. Creators showing the app in use. Paid and organic.
- Reddit. Answering real questions where the problem is discussed. I have run tests across 40 or more accounts.
- Short-form video on the app’s own accounts.
- Founder-led content. The founder on camera or in text.
- Paid social beyond TikTok, once a creative has proven itself.
- App store optimization. Slow, but it compounds.
Tools matter less than founders think. An ad manager per platform. An attribution tool. A spreadsheet for blended CAC, because no dashboard calculates it the way I want. A brief template and a shared folder for creative. That is the stack.
How AI apps changed the job
My first growth role was in 2020 at Neuphony, an EEG neurofeedback wearable that appeared on Shark Tank India. I did design, social, content and organic acquisition. Hardware, with weeks of research before anyone bought.
AI apps are the opposite. See a video, install, try one prompt, decide. The whole funnel happens in minutes.
That changed four things about the role:
- Creative became the main lever. The video is the landing page.
- Activation became the real conversion event. Installs are cheap. Users who get value in session one are not.
- Organic became a channel with a number attached. ZuAI’s $0.02 blended CAC counts paid and organic together. A number like that does not come from paid spend alone.
- Retention became the growth marketer’s problem, not only the product team’s. If the users I bring in leave in a week, my CAC was wrong.
What to look for when hiring one
Questions I would ask any AI app growth marketer, including me:
- What was the blended CAC on the last app you grew, and how did you calculate it?
- How many creatives did you launch last month?
- What did you kill last month, and why?
- Which organic channel have you made work with a number attached?
- Show me a weekly report you sent a founder.
Red flags: awareness talk without a cost per user. Cannot explain platform CPI versus blended CAC. Has only run paid. Has never shot a creative.
The full scope is on my AI app growth marketer page. The ZuAI case study shows what it produced.
What it costs compared with an agency
No figures here, because every arrangement is different. The shape is what matters.
An agency charges a retainer plus a percentage of ad spend. The fee grows with your budget whether or not results do. Creative is usually extra.
An AI app growth marketer charges a flat monthly fee, sometimes with a bonus tied to CAC or user growth. Creative production is part of the work. You pay ad spend directly to the platforms.
At low spend an agency can look cheaper. At high spend the percentage model costs more than a flat fee, and the incentive points the wrong way.
When a founder should not hire one yet
Do not hire an AI app growth marketer if:
- The app does not retain the users it has. Growth spend on a leaky app is money on fire. Fix retention first.
- You have a few hundred users and no idea who they are. Talk to them yourself before paying someone to find more.
- You cannot afford three months of test budget on top of the fee. The first month is mostly losers.
- You want someone to run ads and nothing else. Hire an agency or a freelancer for that.
At that stage the free playbooks and calculators will get you further than a hire. The Goal Digger Club is for founders working this out together. It is early and there is a waitlist.
Past that stage and want to talk? Book a call.
FAQ
Is an AI app growth marketer the same as a performance marketer?
No. A performance marketer runs paid ads and is judged on platform metrics such as cost per install. An AI app growth marketer owns paid and organic together and is judged on blended CAC, activation and retention. Performance marketing is one skill inside the role, and on its own it misses the organic channels that bring the cheapest users.
Can a founder do this job themselves?
Yes, for a while. Many founders I talk to ran the first loop themselves. Research, creative, launch, measure, kill. The limit is time. Done properly the loop takes most of two working days a week, and it competes with product work. Founders usually hand it over when the app has a channel that works and needs someone to scale it.
How long before an AI app growth marketer shows results?
In my campaigns the first four weeks are setup and losing creatives. By week six there is usually one hook or one channel holding a number. By week twelve either something is scaling or the marketing has exposed a retention problem. Anyone promising results in two weeks is guessing or measuring the wrong thing.