iOS应用变现教程：如何两个月做到月入万美元

Frederick James 2026-08-25 20.3 KB 891.4K 2.2K 𝕏 文章列表

作者分享了自己将iOS应用在约两个月内扩展到月收入1万美元的实战经验。内容分为两部分：第一部分讲解应用开发要点，包括瞄准核心人性需求、优先做iOS原生SwiftUI应用、参考成功模式但不抄袭、尽早提交MVP迭代、使用硬性付费墙（推荐RevenueCat/Superwall）、定价策略、优化新手引导流程（目标完成率75%以上、下载付费率8%+）以及App Store审核与页面优化技巧。第二部分开始讲解如何利用Facebook等社交媒体平台的广告算法进行精准投放获客，强调这不是靠运气或病毒传播，而是借助史上最强大的算法，利润率约30%。

iOS 移动应用变现付费墙 App Store优化 SwiftUI RevenueCat Superwall 增长营销创业

# Financial freedom from mobile apps: A tutorial. **作者**: Frederick James **日期**: 2026-08-07T12:47:34.000Z **来源**: [https://x.com/frederickjames/status/2091503229146194334](https://x.com/frederickjames/status/2091503229146194334) --- ![Image](/doc/5143/image/img_016de07473bd.jpg) I scaled my iOS app to $10k/m. • It took me less than ~2 months. • I'm doing ~$1k per day. • On track to hit $30k/m. ![Image](/doc/5143/image/img_f2e37b6e622d.jpg) $11k in actual revenue per month, not MRR. There are 3 things you should know before we begin. 1. This strategy can cost anywhere from $50 to a few thousand $. 2. Success is not guaranteed but this is about as close as it gets. > We're not relying on luck, snake oil or virality. This is not really a guide on mobile apps. It's a guide on leveraging the most powerful algorithms ever created. 3. You are likely to see ~30% profit margins (if solely using paid ads). ## Part One: The App. If you're an appmaxxer skip this section: you'll know most of this. But, if you're new to the app space this is crucial knowledge. > The app ain't special, but it's important. 1. Sell to a core human desire. • If you take nothing else, take this. • Your app should be a painkiller, serving a deep human need. • Status, wealth, money, religion/meaning, beauty, intimacy, addiction. • This is where most founders go wrong. • The app positioning is crucial to enable our marketing. • Everything is symbiotic. 2. At first: build for iOS only. • 70% of app purchases happen on iOS. • The US makes up ~35% of this. • This marketing strategy is optimised for iOS. • If the app does well, then absolutely build it for android, it's free $$$. • But just starting out? Increase your surface area for luck and focus on iOS. • Build native SwiftUI. > The apple design ecosystem is the easiest way to get a premium & beautiful design fast. Liquid glass comes as standard, and OpenAI models in particular, work extremely well with iOS. ![Image](/doc/5143/image/img_fcd0f94b885b.jpg) iOS liquid glass 3. Follow proven model, but do not copy 1:1. • You'll just get your app removed, and it's a bad look. • Whatever you make, make it beautiful. • If you care about longevity, deliver a good UX. > Iterate and improve. I realised a big app was missing a huge and obvious feature. I made my design unique, better, and centered my app around this feature. You'll be surprised how easy this is! 4. Ship the MVP then iterate. • iOS reviews can take weeks. • Submit the first build early on, the absolute MVP. • Then, continue to iterate and improve. • Don't do any risky things in a first build. 5. App Mechanics: Paywall. • I only ship apps with hard paywalls. • Getting downloads is hard enough. • I would not waste attention on trials. • But it can absolutely work. • Use a remote paywall: RevenueCat or Superwall. • This allows you to run AB tests, and update it remotely. • Ensure you stay compliant with Apple. This is the anatomy of a high converting paywall 👇 TLDR: 1. Social proof. 2. Sell impact not product. 3. Explain key features. 4. Clear compliant pricing. 5. Motivational CTA. > **Frederick James@frederickjames**: [原文链接](https://x.com/frederickjames/status/2085709435498807593) > > This paywall structure made me $10,578 in 62 days. > • Average ~10% conversion. > • Studied 100+ paywalls making it. > These are the 5 sections of a high-converting paywall. > This is the one for His 75 #shipaton > > ![Image](/doc/5143/image/img_be1e1edb7105.jpg) 6. App Mechanics: Pricing. • At first, set pricing to what your competitors use. • Later, run AB tests using Superwall or RevenueCat. • It will feel odd at first, but charge "high" prices. • Generally: 7.99/w, 14.99/m and 49.99/y. • Generally: don't push yearly above $50. • There's good data that this increases refunds. 7. App mechanics: Onboarding. • Spend a good amount of time on the onboarding. • The goal here is twofold: 1. Remind the user they have an issue. 2. Explain that your app can solve it. • Mechanics are highly dependent on niche. • So do some research, and see what others do. • Take the core ideas, do not copy it 1:1. • Around 12 screens is directionaly good. • Longer onboarding can: 1. Increase drop off. 2. Increase conversion. • Aim to reduce friction wherever possible. • Use social proof, scientific stats, etc. • Use images over text: people prefer it. • Quizzes can be good: 4 options max. • Explore permission access within the app. • Ensure you measure the funnel. • Mixpanel, PostHog, any will do. > In onboarding, every screen must fight for it's place. There must be a reason for it being there. Always the outcome and not the product. Some excellent onboardings to study: 1. PrayerLock: Absolute masterclass. 2. CalAI: Data driven setup. 3. Feeld & Hinge: Necessary friction done well. 5. Trading212 & Monzo: Very complex apps. 8. App Mechanics: Extra Conversions. First: Try an abandon checkout offers. - They are allowed, but it's up to the reviewer. - It must be VERY CLEAR and explicitly approved. - Generally: if your app is tasteful and premium. - I'd avoid adding one in your first submission. Second: Use App Store retention offers. - Apple will let us show offers to users within the app store. - It's currently in beta, you can request access. Third: Use exit offers. - This is a flow within your app. - User taps cancel sub button. - You show them an offer before taking to app store. - Generally heavily discounted offering. - You can use remote paywalls for this. - Again, keep this compliant with Apple! ![Image](/doc/5143/image/img_396dbb1a7fc3.jpg) 9. App Mechanics: Numbers. 1. Aim for onboarding completion of 75%+ 2. Download to paid: 4% is a start, aim for 8%+. 3. I don't look at paywall conversion: D2P. App Store Approvals [Bonus Tips]. 1. Add a demo video of submission. 2. If it's taking a while, request a call from apple. 3. Be very bold about what the app does. 4. Under no circumstances mislead your users. 5. Don't fight or disagree, unless it's legit. 6. Request an expedited review if it's legit. I have a full guide on getting an app approved 👇 > **Frederick James@frederickjames**: [原文链接](https://x.com/frederickjames/status/2086419927548809285) > > How to get your iOS app approved in under 48 hours. > Apple only wants the user to be happy. If the app tricks them, it'll be rejected. All the ToS revolve around that; it's simple. 10. App Store Page • Title should contain good keywords. • Astro is a good way to find these. • Do not repeat words between: - Title, subtitle & keywords. • Reviews don't matter, unless they're bad. • If you have a low rating, reset in an update. • This only applies to the numeric ratings. • Text reviews stay between updates. • If you have an unfair one, report it to Apple. • Ask your agent to write an official complaint, use data examples where possible. I have a full guide on generating good screenshots 👇 > **Frederick James@frederickjames**: [原文链接](https://x.com/frederickjames/status/2073727713928446362) > These are some I created for Clavicular's lookmaxxing app. ![Image](/doc/5143/image/img_d6b313bcc289.jpg) ## Part Two: The Algorithms In 2007 Facebook Ads was born. This changed marketing forever (and, arguably, the global economy). > Businesses could now show their ads to: single 34 year old men in London earning over £68,000 who are interested in motorbikes and technology. Not just: whoever sees an ad in a newspaper. You need to understand that this is how every social media platform makes most of their revenue. Every second, dollars are being spent on timeline ad space, bidding for our attention. ![Image](/doc/5143/image/img_71201c8863b6.jpg) 98% of Meta's revenue is advertising. The platforms have trillions of terabytes of consumer behaviour data. They pay thousands of extremely intelligent people to build ML systems beyond our wildest imaginations. All we need to do, is plug our app into the system. And watch it print $$$. The method I used was TikT0k ads. But the principles can (and should) be abstracted across platforms, i.e Meta, Snapchat, etc. > Meta is the OG in the space, but I'd argue, today, the TikT0k algorithm is the most sophisticated. If your target user is on TikT0k, I would try this first. 1: Technical Setup. You'll need to connect an MMP to your app. Think of this like a middleman. Sitting between your app and the ad platform. ![Image](/doc/5143/image/img_cbc005659084.jpg) The flow of data. Some good MMPs: Appsflyer & Appstack. I'd personally recommend Appsflyer (not sponsored 🥀) Small apps that use it: TikT0k, Netflix, Pepsi, Barclays. It's free for the first year & then 7 cents per conversion. Appsflyer Setup: 1. Ask your agent to install the SDK with latest docs. Use the SDK integration test to see events as they come in live. You'll want app open and subscribe. You can use server to server: so your app sends data to RevenueCat or Superwall and then postback to the MMP. I didn't do this, but it can be good later on for direct attribution. 2. Once you're seeing sub events in the MMP, connect MMP to the ad platform. You'll want to ensure the events are mapped properly. 3. Now, test the integration by sending events to TikT0k. Once you see a subscribe or app isntall event in your "Events Manager" tab we're good to go. 4. This is specifically for TT, but all platforms behave similarly. ![Image](/doc/5143/image/img_ec8793a03b30.jpg) In app events mapped to TT. 2: Campaign Settings. 1. App Promotion -> App install -> iOS dedicated campaign. 2. Campaign budget -> at least $50/day. I start at $70/day. 3. Goal Based Budget Increase. This setting is what gave me financial freedom. All it does, is increase your daily budget, if the platform gets conversions at your target CPA. Think of this like the snowball effect. If the campaign is doing amazing, you don't want it to stop right? I set it to 20% increase up to 20 times. On $70 of spend this scales to $350/day. I'd recommend keeping it around 20% to 30%. 4. Optimisation Goal: In-app event -> Subscribe OR Install. 5. Campaign Type: There are two choices here. Maximum Results: Will try and spend all of your money, regardless of your target CPA (Cost Per Acquisition). Good for brute force approach. Target Cost Per Result: Will spend slower and maybe not at all, but will try much harder to get you conversions at the CPA you set. 6. Audience Targeting: Switch to "Manual Targeting". Huge thing people miss. If you use automatic, your suggestions are just that: suggestions. US only: US spends the most on iOS, and you don't pay VAT. Many don't realise a subscribe in Europe is like 35% of the actual cost: if a user pays $10 you only get $6.5. Add gender demographics if it matters, and then English speaking: tends to have higher income. Also set age to at least adult ages. Conversion can be higher but so are refunds, personally I wouldn't mess around with this. You can play around with tighter targeting: income, phone type, occupation. Generally I've found it best to let the algorithm decide. 7. Placements: Manual -> TikT0k only. 8. I only use one ad group. And then have ~6 ads within each group. Post the videos on your page first, and then select them within the manager panel. Turn of auto recommendations, choose manually. CTA: Install now, Download now etc - only use ones that make sense. Selling Points: Remove all of these, looks weird. Automatic Enhancements: Turn all these off other than CTA. Destination: I go straight to the app store. Sharing: I allow comments, sharing etc. 3: The Creatives. 1. Don't overthink the actual creatives. You need to think like the ideal user, easiest way is to embed yourself in their world, curate your feed to theirs, scroll and learn the jargon and memes. Save organically viral posts as you go. You'll can tell organic viral because the comment to view ratio will be fairly high. A post with millions of views and only ~10 comments has been paid. 2. Recreate the posts. Seedance is easily the most powerful model available right now. But, what I prefer to do is simply film the video with friends. Takes a bit more effort but generally these convert a bit better I think. Also, platforms are cracking down on AI UGC. 3. Filter comments on your TT account. This is a huge one, you can filter out comments by keywords: paid, cost, money, bad, fake, scam, etc. These comments will definitely have a bad effect on your ROAS. So filter them within your TT account settings. I have maybe 50 words now. 4. Find at least 5 different formats. So, slideshows are one format, UGC hook then app demo is another format. On your first campaign, you want to test out at least 5 different formats. 4: Running Ads & Spending $$$ 1. When I started, I ran 5 formats, with one post each. Five videos total. Looking back, it's better to experiment with more, as while I had a good idea what would work, you really never quite know. 2. TT will try and spend where it thinks it'll get the best returns. Give each post at least $50 of spend, this should be enough to test the creative. In the beginning, the only thing that matters is CPA < LTV: Cost of acquiring the user is less than their lifetime value. At first, you probably won't know your LTV. If you're following all of the above, your LTV will likely be in the $20-$30 region. Conservatively set it to $20. 3. The Learning Phase. If you make a small loss, keep running it. This is crucial for what's called the "Learning Phase". There's a common misconception that this only applies at campaign level. This is incorrect. Your account also has a learning phase. Getting rid of all the specifics, your only job when running paid ads, is to send the ad platform as much signal as possible; signal about what kind of user is most likely to spend money in your app. This creates a feedback loop for the platform. As you spend more money, the targeting becomes more accurate. Eventually, the platform can filter the millions of users scrolling, and show your app to the ones that are extremely likely to buy. 4. If you make a huge loss, stop running the ad. After $50 no conversions at all, stop the ad and reallocate spend. If you get no conversions on any of the formats / posts. They try again from the start. You're taking $50 bets each time. When you hit a winner, you can ride it for weeks, and if you're lucky, months. 5. Once you have a winner (profitable ROAS) scale it. If you're using GBBI, this will be handled for you. If you're using max results, only increase spend by MAX 30% each few days. The platforms themselves recommend this. This is because your campaign has been specifically set up in order to deliver you customers at the variables you set. Change one variable drastically can take days to recover. ## Part Three: Apple Admaxxing I hit $1k/m in 25 days using just apple ads. You get $100 credit FREE from Apple. So, it's a no brainer to try. ![Image](/doc/5143/image/img_4c4fa235780e.jpg) A few words on Apple Ad Maxxing: • I saw a much higher ROAS on Apple Ads than TT. • Apple Ads rely (mostly) search intent. • So, there was a ceiling for scale. • If no one searches for your app, you can't get users. This method will not work for niches where: 1. Ad spend is non viable for indies (i.e finance). 2. Saturated, you'll likely need to loss lead to acquire customers. I have a full guide breaking down the approach 👇 > **Frederick James@frederickjames**: [原文链接](https://x.com/frederickjames/status/2071187278286733629) > The TLDR: • Get attribution running - Apple Ads & RevenueCat/Superwall. • Ensure you have a great app store page (screenshots, etc). • Split campaigns into tiers: - [US], [AU, UK, JP], [TH, DE...] • Bid strategy: Manage Bids. • Search Match: OFF. • Specific keywords on [competitor name]. • Use apple's recommended bid. • Budget doesn't matter: can start $20/day. • More budget = data faster. • If it doesn't spend after 48hrs, scale it 20%. • Export data sets, give to your agent. • Identify winners and add budget. ## Part Four: The Disclaimers 1. Money & profit. Paid ads for mobile apps generally have low margins because we have to account for: • 20% ish Local VAT (if outside US). • 15% Apple store fee. • 1% to RC, Superwall, etc. • That's before your business tax! If you hit ~30% profit margin, I'd call it a success. If you're really cracked you can push this to 40%. The ceiling caps at around 50%, and this is usually reserved for big companies who have fine tuned their funnel on a massive scale. I'd be very skeptical of anyone claiming more than this. There are few things in life that reliably generate a 30% return. And remember, we're we're working 10hrs a week. We put money into the machine and it generates more money. Think of this entire strategy less as a product, and more of an investment. If there was a stock that almost guaranteed 30% profit, you'd buy it right? > Personally, my ~$11k/m, nets me roughly $3,300 in profit each month. While not life changing money, this is financial freedom in a lot of the world. Including SEA where I'm going! I'm also on track for $9,000/m profit as I scale. CPA generally rises with higher budgets, so we'll see how this next step goes. Notable mention is that I've run out of money, so my ads are on pause. I get paid by Apple soon, and that'll get the ball rolling again! This is important because you need cash reserves in order to make this work. 2. App Acquisition If you ever want to sell your app, relying only on paid user acquisition will almost always lower your multiple unless you have many months / years of reliable profit. If you're aiming at a mutli six figure exit, you will want to reduce your exposure to paid marketing. 3. My approach to success. A lot of what I've said is high level and includes a lot of caveats. For example, only building on iOS. Obviously, there are examples of people succeeding only building Android. But I'm not interested in the outliers. My time (and, arguably, yours) should be spent on the highest chance for ROI at any given time. > If you're starting out, I'd advise you to adopt this mindset. If there's good data that x works, try that first, before deviating to y. 4. Absolutist Claims. This is what worked for me (and a lot of other successful appmaxxers). While I've joked on the timeline about paid ads being like day trading, it is a fairly reliable method for growth. But obviously this remains anecdotal for me personally. I strongly believe this pattern is reproducible, and will be trying myself again with another app. That being said, there are infinite ways to win. You should always try anything and everything. Take advice as directional, and test it yourself. It was only through experimenting with the GBBI that I found explosive success. This is a guide on $10k/m. There are levels to this game. $10k/m is an incredible achievement, but I'm sure when I look back at $100k/m I'll change a lot of these views. Try anything and everything. You'll see years of silence, and days of noise Make something beautiful. ## 相关链接 - [Frederick James](https://x.com/frederickjames) - [@frederickjames](https://x.com/frederickjames) - [767K](https://x.com/frederickjames/status/2091503229146194334/analytics) - [Aug 7](https://x.com/frederickjames/status/2085709435498807593) - [#shipaton](https://x.com/hashtag/shipaton?src=hashtag_click) - [21K](https://x.com/frederickjames/status/2085709435498807593/analytics) - [Aug 9](https://x.com/frederickjames/status/2086419927548809285) - [92K](https://x.com/frederickjames/status/2086419927548809285/analytics) - [Jul 5](https://x.com/frederickjames/status/2073727713928446362) - [109K](https://x.com/frederickjames/status/2073727713928446362/analytics) - [Jun 28](https://x.com/frederickjames/status/2071187278286733629) - [725K](https://x.com/frederickjames/status/2071187278286733629/analytics) - [Upgrade to Premium](https://x.com/i/premium_sign_up) - [8:30 PM · Aug 23, 2026](https://x.com/frederickjames/status/2091503229146194334) - [767.9K Views](https://x.com/frederickjames/status/2091503229146194334/analytics) - [View quotes](https://x.com/frederickjames/status/2091503229146194334/quotes) --- *导出时间: 2026/8/25 13:36:22* --- ## 中文翻译 # 通过移动应用实现财务自由：一份教程。 **作者**: Frederick James **日期**: 2026-08-07T12:47:34.000Z **来源**: [https://x.com/frederickjames/status/2091503229146194334](https://x.com/frederickjames/status/2091503229146194334) --- ![Image](/doc/5143/image/img_016de07473bd.jpg) 我把我的 iOS 应用做到了每月 1 万美元。• 只花了不到约 2 个月。• 目前每天收入约 1000 美元。• 有望达到每月 3 万美元。 ![Image](/doc/5143/image/img_f2e37b6e622d.jpg) 每月 1.1 万美元的实际收入，不是 MRR（月度经常性收入）。在开始之前，有 3 件事你应该知道。 1. 这套策略的成本从 50 美元到几千美元不等。 2. 成功并非百分百保证，但这已经是最接近保证的方法了。 > 我们不依赖运气、骗局或病毒式传播。这其实不是一份关于移动应用的指南，而是一份关于利用史上最强大算法的指南。 3. 你的利润率大约在 30% 左右（如果仅使用付费广告）。 ## 第一部分：应用本身如果你已经是应用圈的行家，可以跳过这一节：这些你大部分都知道。但如果你是应用领域的新手，这些是至关重要的知识。 > 应用本身并不特别，但很重要。 1. 面向核心人性欲望销售。• 如果其他都记不住，记住这一点。• 你的应用应该是“止痛药”，服务于深层的人性需求。• 地位、财富、金钱、宗教/意义、美貌、亲密关系、成瘾。• 大多数创始人都在这里犯错。• 应用的定位对于我们的营销至关重要。• 一切都是共生的。 2. 起步阶段：只为 iOS 开发。• 70% 的应用购买发生在 iOS 上。• 其中美国占约 35%。• 这套营销策略是针对 iOS 优化的。• 如果应用表现好，那绝对要为安卓开发，那是白捡的钱。• 但刚起步时？增加你的“幸运接触面”，专注于 iOS。• 使用原生 SwiftUI 开发。 > 苹果的设计生态系统是最容易快速获得高端精美设计的方式。液态玻璃是标配，而且 OpenAI 的模型在 iOS 上表现尤其出色。 ![Image](/doc/5143/image/img_fcd0f94b885b.jpg) iOS 液态玻璃 3. 遵循已被验证的模式，但不要 1:1 照抄。• 否则你的应用会被下架，而且观感很差。• 无论你做什么，都要做得漂亮。• 如果你在乎长期发展，就要提供良好的用户体验。 > 迭代并改进。我发现一个大型应用缺少一个巨大且显而易见的功能。我让我的设计独一无二、更加出色，并围绕这个功能构建我的应用。你会惊讶于这有多容易！ 4. 先发布 MVP，然后迭代。• iOS 审核可能需要数周。• 尽早提交第一个构建版本，绝对的最小可行产品。• 然后持续迭代改进。• 第一个版本不要做任何高风险的事。 5. 应用机制：付费墙。• 我只发布带硬性付费墙的应用。• 获得下载量已经够难了。• 我不会在免费试用上浪费用户注意力。• 但试用也确实可以奏效。• 使用远程付费墙：RevenueCat 或 Superwall。• 这让你可以进行 AB 测试并远程更新。• 确保遵守苹果的规定。这是一个高转化率付费墙的解剖图 👇 要点总结：1. 社会证明。2. 销售效果而非产品。3. 解释关键功能。4. 清晰合规的定价。5. 有激励性的 CTA。 > **Frederick James@frederickjames**: [原文链接](https://x.com/frederickjames/status/2085709435498807593) > > 这个付费墙结构在 62 天内为我赚了 10,578 美元。 > • 平均约 10% 的转化率。 > • 研究了 100+ 个付费墙总结而成。 > 这是一个高转化付费墙的 5 个部分。 > 这是 His 75 的付费墙 #shipaton > > ![Image](/doc/5143/image/img_be1e1edb7105.jpg) 6. 应用机制：定价。• 起初，将定价设为与竞争对手一致。• 之后，使用 Superwall 或 RevenueCat 进行 AB 测试。• 一开始会感觉奇怪，但要收“高”价。• 一般而言：周订阅 7.99 美元，月订阅 14.99 美元，年订阅 49.99 美元。• 一般而言：年订阅不要超过 50 美元。• 有充分数据显示这会增加退款率。 7. 应用机制：新手引导。• 在新手引导流程上花足够多的时间。• 目标有两个：1. 提醒用户他们有一个问题。2. 说明你的应用可以解决这个问题。• 具体机制高度依赖于细分领域。• 所以要做些调研，看看别人是怎么做的。• 借鉴核心思路，不要 1:1 照抄。• 大约 12 个屏幕是不错的方向。• 更长的新手引导可能：1. 增加流失。2. 提高转化。• 尽可能减少摩擦。• 使用社会证明、科学统计数据等。• 用图片代替文字：人们更喜欢图片。• 测验形式可以不错：最多 4 个选项。• 在应用内探索权限请求的时机。• 确保对漏斗进行数据监测。• Mixpanel、PostHog，任何一个都行。 > 在新手引导中，每一屏都必须为其存在而“战斗”，必须有它存在的理由。永远关注结果而非产品。一些值得学习的优秀新手引导：1. PrayerLock：绝对的大师课。2. CalAI：数据驱动的设置。3. Feeld 和 Hinge：恰到好处的必要摩擦。5. Trading212 和 Monzo：非常复杂的应用。 8. 应用机制：额外转化。第一：尝试弃购挽留优惠。- 这是被允许的，但取决于审核人员。- 必须非常清晰明确并获得批准。- 一般而言：如果你的应用有品位且高端。- 第一版提交时我建议避免使用。第二：使用 App Store 留存优惠。- 苹果允许我们在应用商店内向用户展示优惠。- 目前处于测试阶段，可以申请访问权限。第三：使用退出优惠。- 这是应用内的一个流程。- 用户点击取消订阅按钮。- 在跳转到应用商店之前，向他们展示一个优惠。- 通常是大幅折扣。- 可以用远程付费墙实现。- 同样，要符合苹果的规定！ ![Image](/doc/5143/image/img_396dbb1a7fc3.jpg) 9. 应用机制：关键数字。1. 新手引导完成率目标为 75% 以上 2. 下载到付费：4% 是起步，目标是 8% 以上。3. 我不看付费墙转化率，只看 D2P（下载到付费）。 App Store 审核【附加技巧】。1. 提交时附上一个演示视频。2. 如果审核时间过长，要求苹果来电沟通。3. 对应用的功能描述要非常大胆直接。4. 绝对不要误导用户。5. 不要争辩或反驳，除非有正当理由。6. 如果理由充分，可申请加急审核。我有一份完整的获得应用批准的指南 👇 > **Frederick James@frederickjames**: [原文链接](https://x.com/frederickjames/status/2086419927548809285) > > 如何在 48 小时内让你的 iOS 应用获得批准。 > 苹果只想让用户满意。如果应用欺骗用户，就会被拒绝。所有条款都围绕这一点，很简单。 10. App Store 页面 • 标题应包含优质关键词。• Astro 是寻找关键词的好工具。• 不要在以下位置重复用词：- 标题、副标题和关键词。• 评价不重要，除非是差评。• 如果你的评分很低，可以在更新中重置。• 这只适用于数字评分。• 文字评价在更新之间会保留。• 如果有不公正的评价，向苹果举报。• 让你的代理人写一份正式投诉，尽可能使用数据实例。我有一份完整的制作优质截图的指南 👇 > **Frederick James@frederickjames**: [原文链接](https://x.com/frederickjames/status/2073727713928446362) > 这些是我为 Clavicular 的颜值提升应用制作的截图。 ![Image](/doc/5143/image/img_d6b313bcc289.jpg) ## 第二部分：算法 2007 年，Facebook 广告诞生了。这永远地改变了营销（可以说也改变了全球经济）。 > 企业现在可以向以下人群展示广告：伦敦 34 岁的单身男性，年收入超过 68,000 英镑，对摩托车和科技感兴趣。而不只是：任何看到报纸广告的人。你需要理解，这是每个社交媒体平台大部分收入的来源。每一秒，都有美元花在信息流广告位上，竞拍着我们的注意力。 ![Image](/doc/5143/image/img_71201c8863b6.jpg) Meta 98% 的收入来自广告。这些平台拥有数万亿 TB 的消费者行为数据。它们雇用数千名极其聪明的人来构建远超我们想象的机器学习系统。我们所要做的，就是把应用接入这个系统。然后看着它印钱。我使用的方法是 TikT0k 广告。但这些原则可以（也应该）推广到其他平台，如 Meta、Snapchat 等。 > Meta 是这个领域的鼻祖，但我认为，如今 TikT0k 的算法是最先进的。如果你的目标用户在 TikT0k 上，我会先尝试这个。 1：技术设置。你需要将一个 MMP（移动归因平台）连接到你的应用。把它想象成一个中间人，位于你的应用和广告平台之间。 ![Image](/doc/5143/image/img_cbc005659084.jpg) 数据流。一些不错的 MMP：Appsflyer 和 Appstack。我个人推荐 Appsflyer（无赞助 🥀）。使用它的小应用有：TikT0k、Netflix、百事、巴克莱银行。第一年免费，之后每次转化收 7 美分。 Appsflyer 设置： 1. 让你的代理人按照最新文档安装 SDK。使用 SDK 集成测试实时查看传入的事件。你需要 app_open 和 subscribe 事件。你可以使用服务器对服务器模式：应用将数据发送到 RevenueCat 或 Superwall，再回传给 MMP。我没有这样做，但之后用于直接归因可能有好处。 2. 在 MMP 中看到订阅事件后，将 MMP 连接到广告平台。确保事件正确映射。 3. 现在，通过向 TikT0k 发送事件来测试集成。当你在“Events Manager”标签页中看到 subscribe 或 app_install 事件时，就绪了。 4. 这是针对 TT 的具体说明，但所有平台的行为都类似。 ![Image](/doc/5143/image/img_ec8793a03b30.jpg) 应用内事件映射到 TT。 2：广告系列设置。 1. 应用推广 -> 应用安装 -> iOS 专用广告系列。 2. 广告系列预算 -> 每天至少 50 美元。我从每天 70 美元开始。 3. 基于目标的预算增加（GBBI）。这个设置给了我财务自由。它所做的只是：如果平台以你的目标 CPA 获得转化，就增加你的每日预算。把它想象成滚雪球效应。如果广告系列表现出色，你不会想让它停下来对吧？我设置为增加 20%，最多 20 次。70 美元的预算可以扩展到每天 350 美元。我建议保持在 20% 到 30% 之间。 4. 优化目标：应用内事件 -> 订阅或安装。 5. 广告系列类型：有两种选择。最大效果：会尝试花光你所有的钱，不管你的目标 CPA（获客成本）是多少。适合暴力强推。目标成本：花费较慢，甚至可能不花费，但会尽全力以你设定的 CPA 获得转化。 6. 受众定向：切换到“手动定向”。很多人忽略这一点。如果用自动定向，你的建议就只是建议。仅限美国：美国在 iOS 上的消费最多，而且你不用交 VAT。很多人没意识到，在欧洲一次订阅大约会扣掉实际费用的 35%：如果用户支付 10 美元，你只能拿到 6.5 美元。如果有需要可以添加性别人口属性，然后选择英语用户：通常收入更高。年龄至少设置为成年人。转化率可能更高，但退款率也更高，我个人不会在这方面冒险。你可以尝试更严格的定向：收入、手机型号、职业。总的来说，我发现让算法自己决定最好。 7. 版位：手动 -> 仅限 TikT0k。 8. 我只用一个广告组。每个组内放约 6 个广告。先把视频发布到你的主页，然后在管理面板中选择它们。关闭自动推荐，手动选择。CTA：立即安装、立即下载等——只用说得通的。卖点：全部移除，看起来很怪。自动增强：除 CTA 外全部关闭。落地页：我直接跳转到应用商店。分享：我允许评论、分享等。 3：素材。 1. 不要过度纠结素材本身。你需要像理想用户一样思考，最简单的方法是融入他们的世界，把你的信息流调成他们的样子，滑动浏览并学习行话和梗。随手保存自然走红的帖子。判断自然走红的方法：评论与浏览量的比率相当高。一个有几百万浏览量但只有约 10 条评论的帖子是付费推广的。 2. 重制这些帖子。Seedance 无疑是目前最强大的模型。但我更喜欢直接和朋友一起拍摄视频。这需要更多精力，但我认为这些通常转化率更好一些。而且，平台正在打击 AI 生成的 UGC。 3. 在你的 TT 账号上过滤评论。这一点非常重要，你可以按关键词过滤评论：付费、费用、钱、差、假、骗局等。这些评论肯定会对你 ROAS 产生负面影响。所以在你的 TT 账号设置中过滤它们。我现在大概有 50 个关键词。 4. 找到至少 5 种不同的格式。比如，轮播图是一种格式，UGC 开场然后应用演示是另一种格式。第一个广告系列中，你要测试至少 5 种不同格式。 4：投放广告和花钱 1. 我开始时投放了 5 种格式，每种一个帖子，总共 5 个视频。回头看，多测试一些更好，因为虽然我对什么有效有不错的判断，但你永远无法完全确定。 2. TT 会尝试把它认为回报最好的地方花预算。给每个帖子至少 50 美元的预算，这应该足够测试素材了。开始阶段，唯一重要的是 CPA < LTV：获取用户的成本低于其生命周期价值。起初，你可能不知道自己的 LTV。如果你遵循以上所有建议，你的 LTV 可能在 20-30 美元区间。保守地设为 20 美元。 3. 学习期。如果小亏，继续投放。这对所谓的“学习期”至关重要。常见的误解是这只适用于广告系列层面。这是错误的。你的账户也有学习期。抛开所有细节，投放付费广告时你唯一的任务，就是向广告平台发送尽可能多的信号；关于什么样的用户最可能在你的应用中花钱的信号。这为平台创建了反馈循环。随着你花更多钱，定向会变得更精准。最终，平台可以从数百万滚动浏览的用户中筛选，把你的应用展示给那些极有可能购买的人。 4. 如果亏损严重，停止投放。花费 50 美元后仍无任何转化，停掉该广告并重新分配预算。如果任何格式/帖子都没有转化，就从头再来。你每次是在下 50 美元的注。当你命中赢家时，可以吃它几周，幸运的话几个月。 5. 一旦找到赢家（盈利的 ROAS），就扩大规模。如果你使用 GBBI，这会自动处理。如果你用最大效果模式，每几天最多增加 30% 的预算。平台自己也这么建议。这是因为你的广告系列是专门按照你设置的变量来为你带来客户的。大幅改变一个变量可能需要数天才能恢复。 ## 第三部分：苹果广告极致玩法我只用苹果广告就在 25 天内达到了每月 1 千美元。苹果免费送你 100 美元额度。所以，尝试一下是毫无悬念的选择。 ![Image](/doc/5143/image/img_4c4fa235780e.jpg) 关于苹果广告的几点说明：• 我在苹果广告上看到的 ROAS 比 TT 高得多。• 苹果广告（主要）依赖搜索意图。• 所以规模存在天花板。• 如果没人搜索你的应用，你就得不到用户。这种方法在以下细分领域行不通：1. 广告支出对独立开发者不可行的领域（如金融）。2. 饱和市场，你可能需要亏本获客。我有一份完整拆解该方法的指南 👇 > **Frederick James@frederickjames**: [原文链接](https://x.com/frederickjames/status/2071187278286733629) > 要点总结：• 搭建归因 —— Apple Ads 与 RevenueCat/Superwall。• 确保你有出色的应用商店页面（截图等）。• 将广告系列分层：- [美国]、[澳、英、日]、[泰、德……] • 出价策略：管理出价。• Search Match：关闭。• 在 [竞争对手名称] 上做精准关键词。• 使用苹果建议的出价。• 预算不重要：每天 20 美元即可开始。• 预算越多 = 数据越快。• 如果 48 小时后未消耗，提高 20%。• 导出数据集，交给你的代理人。• 识别赢家并增加预算。 ## 第四部分：免责声明 1. 钱和利润。移动应用的付费广告通常利润率较低，因为我们必须计入：• 约 20% 的当地 VAT（如果在美国以外）。• 15% 的苹果商店佣金。• 1% 给 RC、Superwall 等。• 这还没算你的企业税！如果你的利润率达到约 30%，我会称之为成功。如果你真的很强，可以推到 40%。上限大约在 50%，这通常属于那些在大规模上精细调优漏斗的大公司。我会对任何声称超过这个数字的人持高度怀疑态度。生活中很少有东西能稳定产生 30% 的回报。记住，我们每周只工作 10 小时。我们把钱投进机器，它产生更多的钱。把这套策略更多看作一项投资，而不是一个产品。如果有一只几乎保证 30% 利润的股票，你会买对吧？ > 就我个人而言，每月约 1.1 万美元，每月净赚大约 3,300 美元利润。虽然不是改变人生的钱，但在世界很多地方这就是财务自由。包括我即将去的东南亚！随着规模扩大，我也有望达到每月 9,000 美元利润。CPA 通常会随预算增加而上升，所以让我们看看下一步如何。值得一提的是，我的钱已经用完了，所以广告暂停了。苹果很快会给我打款，到时候就会重新滚动起来！这一点很重要，因为你需要现金储备才能让这套方法运转。 2. 应用出售如果你想卖掉你的应用，仅依赖付费用户获取几乎总会降低你的估值倍数，除非你有多个月/多年的稳定利润。如果你的目标是六位数以上的出售退出，你需要减少对付费营销的依赖。 3. 我对成功的态度。我说的很多内容都是高层次的，并附带很多前提。比如，只在 iOS 上开发。显然，也有只在安卓上开发成功的例子。但我对离群值不感兴趣。我的时间（可以说你的时间也是）应该花在任何时刻 ROI 概率最高的地方。 > 如果你刚起步，我建议你采纳这种心态。如果有充分数据显示 x 有效，先尝试 x，再去偏离尝试 y。 4. 绝对化的说法。这是对我（以及许多其他成功的应用玩家）有效的方法。虽然我在时间线上开玩笑说付费广告像日内交易，但它是一个相当可靠的增长方法。但显然，对我个人而言这只是个人经验。我坚信这个模式是可复现的，我会用另一个应用再次尝试。话虽如此，成功的方法有无数种。你应该始终尝试任何可能性。把这些建议当作方向性的参考，自己去测试。正是因为试验了 GBBI，我才获得了爆发式的成功。这是一份关于每月 1 万美元的指南。这个游戏有很多层级。每月 1 万美元是一个了不起的成就，但我敢肯定，当我回头看每月 10 万美元的时候，我会改变很多这些观点。尝试任何可能性。你会经历多年的沉寂，和几天的喧嚣。做一些美丽的东西吧。 ## 相关链接 - [Frederick James](https://x.com/frederickjames) - [@frederickjames](https://x.com/frederickjames) - [767K](https://x.com/frederickjames/status/2091503229146194334/analytics) - [Aug 7](https://x.com/frederickjames/status/2085709435498807593) - [#shipaton](https://x.com/hashtag/shipaton?src=hashtag_click) - [21K](https://x.com/frederickjames/status/2085709435498807593/analytics) - [Aug 9](https://x.com/frederickjames/status/2086419927548809285) - [92K](https://x.com/frederickjames/status/2086419927548809285/analytics) - [Jul 5](https://x.com/frederickjames/status/2073727713928446362) - [109K](https://x.com/frederickjames/status/2073727713928446362/analytics) - [Jun 28](https://x.com/frederickjames/status/2071187278286733629) - [725K](https://x.com/frederickjames/status/2071187278286733629/analytics) - [Upgrade to Premium](https://x.com/i/premium_sign_up) - [8:30 PM · Aug 23, 2026](https://x.com/frederickjames/status/2091503229146194334) - [767.9K Views](https://x.com/frederickjames/status/2091503229146194334/analytics) - [View quotes](https://x.com/frederickjames/status/2091503229146194334/quotes) --- *导出时间: 2026/8/25 13:36:22*