There are roughly 2,000 new apps released on the App Store every day. Apple Ads charges $10–20 per paying user in competitive categories. And Apple just demoed on-device food recognition in iOS 27’s camera app, wiping out the differentiation of every calorie photo app in one keynote slide. The follow-up Reddit thread, “Why would anyone start a new app today?”, is the indie community asking out loud whether the math still works.

This post is the playbook I am using to answer that question in 2026. The economics are real. The way through them is not the one Apple’s marketing materials describe.

Stop treating the App Store as a discovery engine

The App Store is a distribution channel, not a discovery channel. People download apps they have already heard of elsewhere. The App Store turns intent into install. It does not create intent.

• ASO matters only for converting intent into install, not for driving it.
• Featuring is a bonus, not a plan.
• The category leaderboards reward apps that already have a flywheel.

This has been the reality since around 2017. In 2026, it is the only reality.

Pick a niche Apple will not Sherlock

Sherlocking is Apple absorbing third-party functionality into the OS. The signals for lower Sherlocking risk:

• Apple shipped a comparable system feature five years ago and you can still differentiate. QR scanners, PDF scanners, weather apps — Apple has system versions, third-parties still thrive. The differentiation is depth or niche use case.
• Your domain requires data Apple does not have. Specialty cooking apps, regional cuisines, professional reference, vertical B2B, hobbyist hardware integrations. Apple’s on-device food model doesn’t know what nasi lemak looks like.
• Your domain requires regulatory accountability. Medical, accounting, legal, aviation. Apple will not ship these into the OS because they would inherit liability.

If your differentiation is “a slightly better UI on top of a feature Apple could ship in any release”, you are on borrowed time.

Build a direct channel before you build the app

Every indie app that has launched well in the last three years had a list before launch. Newsletter, Discord, subreddit, Twitter following, niche forum presence, podcast audience.

The order that works:

1. Identify a community where your target user already gathers.
2. Show up there as a peer for six to twelve months before the app exists.
3. Open a waiting list early. A landing page with a single email field on your own domain. Use Buttondown, ConvertKit, or Beehiiv — services that let you export the list.
4. Convert the list at launch via a direct App Store link with a campaign token:

https://apps.apple.com/app/idXXXXXXXXX?pt=YOURTEAM&ct=launch-email&mt=8

The pt (provider token) and ct (campaign token) surface in App Analytics. Most developers forget this exists.

Run paid ads only against a known funnel

Apple Search Ads at $10–20 per paying user is fine if your LTV is over $100 and you have a working onboarding. It is a runway-incinerator if you don’t.

Before you spend on paid acquisition:
– Know your conversion rate from install to paying customer.
– Know your LTV across at least 90 days via App Store Connect subscription metrics.
– Test creative against a small budget first. $500 across three variants.

Use Apple Ads’ advanced campaign type for keyword-level data. The basic type is fully automated and you cannot debug it.

What works that doesn’t cost $10/install

• Podcast read sponsorships in niche shows. A vertical podcast with 5,000–20,000 listeners charges $300–800 per episode and converts at rates Apple Ads cannot touch.
• YouTube reviews from second-tier creators. The 10k–100k subscriber tier is reachable, will take a free license + small fee, and videos accumulate views for months.
• Content marketing aimed at the long tail of search. A blog post on your domain ranking for “how do I X” in your niche delivers low-volume but very high-intent traffic.
• The single launch-week press push. A short factual press release to five to ten outlets the day before launch. Half will ignore you. One or two will write a paragraph worth more than a month of paid ads.

Accept Sherlocking risk explicitly in the plan

The Sherlocked path that works:
– Go pro-vertical. When Apple Sherlocked simple QR scanners, surviving apps went enterprise.
– Go cross-platform. A Mac + iOS + web product is harder for Apple to fully replicate.
– Sell the company before the next keynote if the moat is thin.

A realistic launch budget

For a solo indie launching a non-Sherlock-prone niche app in 2026:

• Pre-launch community time: 6–12 months part-time. The biggest cost is calendar time.
• Landing page + email service: $20/month.
• First paid ad budget: $500 across three creatives, only after conversion is measured.
• Tools: $0–200/month depending on volume.

Total cash out for the first six months: under $5,000 if you do the community work. The economics are brutal for the commodity case and survivable for the niche case. The whole game is being honest about which one you are in.

---