App growth marketing is the work of getting users into an app, getting them to stick, and doing it at a cost the business can afford, over and over. It covers organic and paid channels, creative, and measurement. This guide is what I use with founders, built from taking ZuAI to 2 million users.
What app growth marketing is
Most marketing is about attention. Mobile app growth marketing is about a number: users added, what each one cost, and how many are still there in 30 days. Everything else serves that number.
The job has four parts:
- Find the channels where your users already are.
- Make creative that gets them to install.
- Measure what each user cost and whether they stayed.
- Kill what does not hold its number. Scale what does.
That is the whole discipline. The rest is detail.
The app funnel
A website funnel ends at a form. An app funnel keeps going after the install, and the later stages decide whether the early ones were worth paying for.
- Install. The download. The cheapest step and the one most teams celebrate too early.
- Activation. The user reaches the first real moment of value. For an AI app that is usually the first useful output. Define this event before you spend a dollar.
- Retention. The user comes back. I track day 1, day 7 and day 30 by weekly cohort.
- Referral. The user brings someone else. For consumer apps that is usually a screenshot or a video, not a referral code.
- Revenue. The user pays, upgrades, or watches an ad. This sets the ceiling on what you can spend to get them.
Work it backwards. Revenue per user sets the CAC you can afford. Retention tells you whether installs are worth anything. Activation tells you whether the creative promised the right thing. Then install volume matters.
Organic versus paid
Organic is users who found you without a paid click. Paid is users you bought. Both cost money. Organic costs time, creator fees and tools. Paid adds ad spend on top.
The mistake is treating them as separate. In my campaigns they feed each other. An organic TikTok video that gets traction becomes a paid ad. A paid ad that works reveals a hook for the organic account. A Reddit thread that sends installs tells you what the ad should say.
The number that keeps them honest is blended CAC: all spend, paid and organic, divided by all new users. ZuAI’s $0.02 is a blended number. Neither side alone gets there.
Start organic if you have time and no money. Paid if you have money and no time. Run both as soon as you can, measured as one.
The channels that work for apps in 2026
The channels I run, in the order I usually test them. Most apps need two, not six.
TikTok UGC. Real people using the app on camera. The same footage runs as organic posts and as paid ads. It needs creative in volume, because most videos lose. Details on the TikTok UGC page.
Reddit. People describe their problem in full sentences. Answer the question, mention the app once, and the comment sends installs for months. Pitch and it is gone in an hour. I have tested this across 40 or more accounts. See Reddit marketing.
Short-form video on your own accounts. TikTok, Reels, Shorts. Same footage, cut for each. The account becomes an asset that lowers CAC over time.
Founder-led content. The founder explaining the problem, on camera or in text. People trust a person more than a logo. See founder-led content.
Paid social. TikTok Ads, Meta and the rest. Best used to scale creative that has already worked organically. Worst used as the first thing you try.
App store optimization. Screenshots, title, keywords, reviews. Slow. Compounds. Do it once properly and revisit each quarter.
How to set up measurement
Set this up before the first dollar goes out. Three numbers, read weekly.
Blended CAC. All acquisition spend divided by all new users in the week. Include ad spend, creator fees and tools. Do not use platform CPI. It ignores organic and hides fees.
Payback. Weeks until revenue from a cohort covers what it cost to acquire. If you do not know revenue per user yet, use a target and revise it.
Retention cohorts. Group users by install week. Track the share still active at day 1, day 7 and day 30. A channel that brings cheap installs with bad retention is not cheap.
What you need:
- An attribution tool that connects installs to campaigns.
- An analytics event for activation, defined before launch.
- A spreadsheet with one row per week: spend, new users, blended CAC, activation, retention.
That is enough. The free growth calculators cover the arithmetic.
A 90 day plan
This is the plan I run with founders. It assumes the app already retains some users.
Weeks 1 to 2: setup.
- Define activation. One event.
- Set up attribution and the weekly spreadsheet.
- Set the CAC ceiling from revenue per user.
- Read 50 reviews and 20 Reddit threads. Write down the exact words users use.
- Brief the first 20 creatives from those words.
Weeks 3 to 6: test.
- Launch across one organic and one paid channel.
- Fixed spend per creative. Read the numbers only at the threshold.
- Kill on the same day every week.
- Replace every killed creative with a new one. Change the hook, not the caption.
- By week 6 you want one hook or one channel holding a number two weeks in a row.
Weeks 7 to 12: scale.
- Raise budget on the winner in steps, not jumps. If CAC drifts at 3 times the budget, it was not ready.
- Copy the winning hook into new formats and new creators.
- Add the second channel using the same words.
- Keep testing new creatives at the same volume. Winners decay.
- By week 12 either something is scaling or the app has a retention problem. Both are useful to know.
Common mistakes
The mistakes I see most on founder calls:
- Measuring installs instead of activation. Cheap installs that never activate are the most expensive users you will buy.
- Running paid before organic has proven a hook. You pay to learn what Reddit would have told you for free.
- Testing 3 creatives and calling it a test. Three is not a test. Twenty is a start.
- Killing too slowly. A losing creative that runs another week is a week of budget gone.
- Scaling too fast. Doubling budget in a day breaks the number that made you want to double it.
- Reporting platform CPI. It flatters every channel.
- Outsourcing creative to people who have never used the app. The video is the landing page.
When to bring in help
Do the first loop yourself if you can. You will learn what your users say and what your creative has to promise. Nobody can shortcut that.
Bring in help when:
- One channel is holding a number and you do not have the hours to scale it.
- Creative volume is the bottleneck and you cannot brief, shoot and cut 20 a month alone.
- Blended CAC is drifting and you cannot see why.
- You are spending real money and nobody owns the whole number.
What that help looks like is on the app growth marketing page. The ZuAI case study shows what the loop produced. The Goal Digger Club is for founders running it together, early, with a waitlist. Or book a call.
FAQ
What is the difference between app growth marketing and app marketing?
App marketing usually means promotion: ads, the store listing, launch PR. App growth marketing covers the whole funnel, from install through activation, retention and revenue, and is judged on cost per user who stays. It includes the promotion. It also includes killing channels that produce installs but not users, which is where most money is saved.
How much should I spend to test app growth marketing?
Enough to give 20 creatives a fair read across 4 weeks. The figure depends on your platform and your CAC ceiling, so I will not quote one. The rule is that test budget must be money you can lose. In my campaigns the first month is mostly losers, and that is the point.
Can app growth marketing work without paid ads?
Yes, and for many apps it should start that way. TikTok UGC, Reddit and founder-led content all bring users without ad spend. The cost is time and creator fees, which still count in blended CAC. Paid is best added once organic has shown which hook works. Then paid buys volume for a message you already know converts.