Niche-Down Deep Dive — use cases, personas, and what culture actually buys

📚 Superseded — this document has been merged. Everything here now lives in The Niche-Down Field Guide, together with the other research docs, deduplicated and reorganised. This page is kept so existing links keep working; it is a snapshot and will not be updated further.

---

Companion to the playbook · 2026-08-11 · Source: AppKittie, Apple App Store, pulled today. Estimated monthly figures, App Store IAP only.

Three things came out of this sweep that change the shape of the last document:

1. Revenue per download is the whole game, and it splits along one line: coaching beats logging. Same behaviour, same category, 10–15× difference.
2. Parenting is the strongest persona category I have found — better $/DL than fitness. I got this wrong on the first pass and corrected it; the story of that mistake is instructive.
3. The culture axis is real but conditional. Christian devotional and Islamic prayer are the same daily behaviour with the same cadence. One earns 4.50–10 per download, the other $0.73.

---

1. The master table — revenue per download

This is the number that decides whether a solo founder can win, not monthly revenue. Everything below is from today's pull.

Read the two ends of that table against each other, because it is the entire strategy:

Ladder $20.06 vs Hevy $2.00. Both are strength training. Both are excellent, highly-rated, daily-use apps. Ladder tells you what to lift; Hevy records what you lifted. Ten times the revenue per user for the app that decides.

Same pattern everywhere. Runna (16.67)coachesrunning; Strava(5.34) records it. MacroFactor ($30.01) tells you what to eat based on your data; a plain calorie log is worth 2–4.Huckleberry(8.00) tells you when to put the baby down; PiyoLog ($0.40) lets you write down when you did.

Rule 1: never build the logger. Build the thing that decides. Logging is a commodity that trends to free. Judgement is what people pay for, and it is also what is genuinely hard — which is why it stays defensible.

---

2. Parenting — the strongest persona category here, and how I nearly missed it

I first searched title-scoped for baby tracker feeding sleep newborn toddler. It returned 5,595 apps whose ceiling was $20,000/month, with dozens of identical zero-revenue clones. The obvious conclusion was that parenting is a dead free-utility space.

That conclusion was wrong, and I checked it because it looked too dead to be true. Searching baby alone surfaced what the narrow query had filtered out:

Four apps between $500k and $1M a month at $8–12.50 per download. That is better per-user economics than almost anything in fitness, on a behaviour with higher cadence (8–12 feeds a day) and vastly higher anxiety.

The clones die and the branded ones print because of Rule 1. "Record the feed" is free. "Your five-month-old is in a leap, here is why she stopped sleeping" is $8–12 a download.

A methodological warning worth keeping: a title-scoped search over the commodity phrasing of a category returns the commodity apps. The winners name themselves after the answer ("Solid Starts", "Napper", "The Wonder Weeks"), not the mechanism ("baby feeding tracker"). Always re-run a category search on the bare noun before concluding a space is dead.

---

3. Faith — where culture pays and where it does not

This is the cleanest natural experiment available on your culture-niching question. Prayer is a daily, identity-driven behaviour in both traditions. Islamic practice has higher cadence — five prayers a day, plus Ramadan.

Christian devotional runs 1.67–10.02 per download. The entire Islamic field outside Muslim Pro tops out at $91k/month, and Muslim Pro itself — 570,662 downloads a month, one of the highest download counts anywhere in this study — earns $0.73 per download.

The cause is not devotion, it is geography and norms: the Muslim audience concentrates in Indonesia, Pakistan, Bangladesh, Egypt and Nigeria, where iOS share and ARPU are a fraction of the US, and where religious content carries a strong cultural expectation of being free.

Rule 2: a culture niche pays only in proportion to where that culture's population sits. The axis is not "which culture," it is "which culture, in which market." Latino-American, Indian- American, Chinese-American and Muslim-American audiences are all high-ARPU; the same cultures in their home markets are not.

Two further notes from this sweep:

• Faith apps live in Reference, not Health & Fitness. Hallow, Bible Chat, Muslim Pro and Abide are all Reference. It is an underrated, much less crowded shelf.
• Bible Note: Sermon Notes (Mar 2025, $208k/month) is a textbook recent niche-down: note-taking → sermon notes. Same move as MenuFit, different category.

---

4. Health & Fitness — the use-case map

Sorted by revenue, the top of the category divides into eight distinct behaviours. This is the menu of bases you can niche from.

The two open lanes are calorie/macro tracking and strength programming, and they are open for the same reason: the behaviour is genuinely daily, the incumbents are generic by design, and the wedge can be a person, a diet, a condition, or an occasion.

---

5. Education and Food & Drink — where the recent wins sit

Education splits into three monetising behaviours, all daily:

• Spoken-English coaching for non-native speakers — YayTalk $535k, LangLearn $506k, BetterSpeak $229k. Three apps above $200k/month, all under two years old, all aimed at Asian learners and priced globally. This remains the single most repeatedly-proven wedge in the study.
• Exam prep — Astra AI $600k. Time-boxed but intensely daily while it lasts. Note NCLEX 2026 at $10k/month on 3,310 downloads: tiny, but that is one person's income from one exam.
• Speech and articulation — Wellspoken $215k at $4.01/DL, released Oct 2025.

Food & Drink is the newest and least defended:

• mise — meal planner — released Dec 2025, already $482k/month at +143% MoM.
• The recipe-management cluster — Miso $84k, CookShelf $74k, Clove $55k, CookGo $50k — is all small teams, all launched in the last year, all still growing.

And Solid Starts sits in Education, not Food & Drink, which is worth noticing: it is a food product that succeeded by being categorised as teaching a parent rather than managing recipes. Shelf choice is a lever.

---

6. Revised candidate shortlist

Reordered on the evidence above. Still hypotheses — run the §7 screen in the playbook before committing to any of them.

Tier 1 — best combination of $/DL, cadence and open space

1. Baby feeding or sleep × a cultural or dietary tradition. Parenting proved $8–12.50/DL and Solid Starts proved that "what do I feed them" is worth 1M/month.EveryoneoftheseproductsisbuiltonWesternweaningnorms.FirstfoodsforChinese, Indian, orhalalhouseholds—soldtodiasporaparentsintheUS/UK/AU—sitsattheintersectionofthehighest−/DL category and the only version of the culture axis that pays. This is now my top recommendation, ahead of the cuisine calorie tracker from the last doc, because the anxiety is higher and the proven per-user economics are better.
2. Strength programming × a persona. Ladder at $20.06/DL is the ceiling to aim at, and the distance to Hevy at $2.00 is the entire argument for being opinionated. Men 50+ remains open and reuses your FiftyStrong machinery wholesale.
3. Spoken English × a professional occasion. Three separate proofs above $200k/month. Medical English for nurses and accent work for Cantonese/Mandarin speakers are the sharpest unclaimed cuts, and the buyers are credentialed professionals rather than casual learners.

Tier 2 — real, with a specific caveat

4. Calorie tracking × cuisine. Still a genuine gap — no cuisine-specific tracker of any scale exists. But treat that emptiness carefully: my search found zero revenue-generating cuisine-specific nutrition apps, and absence of competition is ambiguous evidence. It could mean nobody has tried, or that the food-database build cost exceeds what the niche returns. Validate with a painted door before building a database.
5. Perimenopause cycle tracking. Flo and Clue own general cycle tracking, but both are aimed at menstruating and TTC users; the 45–55 transition is thin. Adjacent to everything you already know about the FiftyStrong audience.

Tier 3 — avoid, with reasons

• Islamic / lower-ARPU-culture apps. High cadence, deep engagement, $0.73/DL. Do not confuse usage with revenue.
• Any pure logger — baby feeding logs, gym logs, habit trackers. Rule 1.
• Meditation, activity-social, general cycle tracking. Brand- or network-dominated; entry cost exceeds what a solo founder can carry.

---

7. The three rules, stated plainly

1. Build the decider, not the logger. Ladder $20.06 vs Hevy $2.00; Huckleberry $8.00 vs PiyoLog $0.40. The app that tells you what to do is worth ~10× the app that records what you did.
2. A culture niche pays in proportion to where that culture's population sits. Christian US 4.50–10/DL vs Islamic global $0.73/DL, on the same behaviour at lower cadence. Target diaspora communities in high-ARPU markets, not home markets.
3. Cadence is necessary but not sufficient — you also need anxiety or identity. Baby feeding is the highest-cadence behaviour in this study and the generic version earns nothing. Prayer is identity. Baby sleep is anxiety. Gym logging is neither, and it is priced accordingly.

---

Caveats

• All figures are third-party estimates covering App Store IAP only. Web-funnel businesses are invisible here — the same blind spot that makes Reverse Health look like a $12k/month company.
• $/DL is a ratio of two estimates, so its error compounds. Use it to sort into bands (under $2 / $2–5 / $5–12 / $12+), not to compare $8.00 against $8.75.
• Revenue per download is not margin. MacroFactor's $30.01 reflects a serious, expensive product with real content costs behind it. High $/DL usually means high cost to deliver — which is precisely the FiftyStrong constraint, restated.
• Search results depend on phrasing, as §2 demonstrates at some cost. Before concluding a category is dead, search the bare noun as well as the descriptive phrase.
• Nothing here is validated demand. It is a map of where to point a priced paywall.