The app ideas worth building are searches people already make on the App Store that no big app owns yet. On 1 November 2026, we measured 585 searches on the US App Store the way ASO Gauge measures a keyword: how few letters people type before the App Store suggests the phrase, which says how much it is searched, and how many ratings the first apps have, which says how hard it is to get in. 353 of those searches are made often. For half of them, the median app of the first 10 has more than 9,000 ratings: years of downloads ahead of any new app. Below are 30 where it has fewer than 1,000, grouped by kind, each with its numbers, then the popular searches that look like good ideas and are already taken.
What decides is less the idea than the words. People who type “guitar tuner” find apps whose median has 76,382 ratings; people who type “drum tuner” find a median of 172.
The 30 app ideas at a glance
A popularity of 41 means the App Store suggests the phrase after five letters or fewer, near the top of its list; a median under 1,000 is what ASO Gauge calls low or medium competition. How the 30 were chosen, and what was left out, is explained further down.
Why the words matter as much as the idea
Four pairs from the same night, each a close need written two ways:
The pairs are close, not identical: a drum is not a guitar, and saying what is wrong with a plant is not saying what plant it is. That is the point. The broad words are held by apps with years of ratings; the narrower words are still searched, and the apps that answer them are small. A new app gets in through the specific words first, then earns the broad ones. The keyword research guide shows how to find such words for an app you already have.
AI app ideas
- AI outfit changer. Suggested after “ai o”, 2nd; popularity 69; median 1. The biggest of the first 10 has 60 ratings, and 5 of the 10 came out in the last 12 months. People want to see clothes on a photo of themselves, and the searches around it are as open: “ai outfit maker” (65, median 6), “ai outfit planner” (51, median 1), “ai outfit generator” (53, median 36).
- AI plant doctor. “ai pl”, 3rd; 55; median 0, and 5 of the first 10 are new. PictureThis, with 1,121,756 ratings, appears in the list, but the six apps named as plant doctors, AI Plant Doctor, PlantFix and Sherleaf among them, have between 0 and 16 ratings. Naming a plant is taken; saying what is wrong with a sick one, from a photo, is not. “ai plant health scanner” (45, median 2) is the same need.
Health and mental health app ideas
These are apps to track and to write, not to diagnose. We left out the searches for tests and symptom checkers, and Apple warns about that side of health:
Medical apps that could provide inaccurate data or information, or that could be used for diagnosing or treating patients may be reviewed with greater scrutiny.
- Anxiety tracker. “anx”, 4th; 65; median 320. Besides Daylio (62,299 ratings), StressWatch (24,874) and Rootd (10,552), the anxiety trackers in the first 10 have between 0 and 323.
- Anxiety journal. “anx”, 8th; 57; median 0, with 5 of the first 10 new. Clarity (29,452), Daylio and Reflectly (81,678) hold three places; the five apps named as anxiety journals all have 0 ratings.
- ADHD timer. “adh”, 7th; 59; median 6, with 5 new. Time Timer (14,614) and Focus Keeper (31,733) are general timers; the apps built around time blindness have 0 to 12 ratings. Apple names simple timers among the kinds of apps it won’t accept unless they are meaningfully different, so this one has to be about ADHD, not about counting down.
- ADHD planner. “adhd planner free”: “adh”, 9th; 55; median 230. Structured (168,267) and Tiimo (21,278) lead; InControl, Planwoo, Doot and Anchor have between 1 and 391. “adhd calendar” (41, median 217) has the same shape.
- ADHD sound app. “adh”, 6th; 61; median 353. Moongate (60,879), Endel (33,971) and Brain.fm (5,554) lead; the two apps actually called ADHD Sound App have 559 and 147.
- Symptom log. “sympt”, 4th; 41; median 3, with 5 new. Bearable (6,585) is the only app with real numbers; the rest have 0 to 972.
- Posture workout. “postu”, 4th; 41; median 2. Bend (174,725) is a stretching app; the posture apps have between 0 and 342 ratings, and one was last updated in 2021.
App ideas where the apps already charge
In each of these searches, at least four of the first 10 apps sell in-app purchases, and they still have few ratings.
- Tip tracker. “tip”, 5th; 63; median 537. For servers and other tipped workers. ServerLife (9,939) and TipTracker (9,464) lead, the rest have 2 to 3,100, and 7 of the first 10 sell in-app purchases.
- Subscription tracker. “subs”, 6th; 49; median 195. Rocket Money (403,866) holds first place, but the dedicated trackers below it have 0 to 8,028 ratings, and one was last updated in 2022.
- Savings goal tracker. “savi”, 9th; 43; median 65. Apart from Oportun (289,200), a savings and loans app, the most rated is Loot (9,112), last updated in 2021.
- Debt snowball. “debt”, 9th; 43; median 698. A payoff plan by the snowball method: Debt Payoff Planner (7,401) ranks first, EveryDollar (84,192), a budget app, third, and the app called Debt Snowball (1,395) was last updated in 2023. “debt free” (41) has a median of 0.
- Inventory tracker. “inve”, 5th; 51; median 220. For small shops and resellers: Sortly (9,555) leads, and one of the first 10 was last updated in 2023. “inventory management” (43, median 202) is the same need.
- Allowance tracker. “allo”, 9th; 43; median 16. Parents tracking pocket money: besides Chores & Allowance Bot (10,960), the first 10 have 0 to 581 ratings, and the app called Allowance Tracker was last updated in 2023.
Simple app ideas for a first app
Small apps, a few screens each, in searches with room.
- Gratitude jar. “grat”, 6th; 49; median 0. The biggest of the first 10 has 402 ratings. One note a day, a jar that fills: a first app can be this small.
- Recipe notes. “reci”, 10th; 41; median 1. Private recipe notes, not a recipe site: the leader has 413 ratings, and two of the first 10 have not been updated in years.
- Chore chart for kids. “chore”, 4th; 41; median 92. Chorsee (11,981) and Chores & Allowance Bot (10,960) lead; the rest have 0 to 932.
- Pantry inventory. “pant”, 6th; 49; median 59. What is in the cupboard: Cooklist (11,370) leads, and the pantry apps have 0 to 1,575 ratings.
- Periodic table. “peri”, 7th; 47; median 113. A study tool: the leader has 5,447 ratings, and the next ones 2,750 and fewer.
- Macro-friendly food. “macr”, 8th; 45; median 0. Meals that fit your macros, rather than another way to count them: the biggest of the first 10 has 649 ratings, and two were last updated in 2024 or before. Counting is taken: “calorie tracker” has a median of 135,541.
App ideas for hobbies, style and the outdoors
- Drum tuner. “dru”, 6th; 61; median 172. The drum tuners among the first 10 have between 0 and 2,336 ratings (iDrumTune), and two of the first 10 are paid apps.
- Piano tuner. “pia”, 10th; 53; median 220. For people who tune pianos: the first 10 have 2 to 10,618 ratings, and two were last updated in 2019 and 2021.
- Piano for kids. “piano kids”: “pian”, 6th; 49; median 652. Baby Piano (384,627) leads, and two of the first 10 were last updated in 2021 and 2024.
- Mushroom identification. “mush”, 4th; 53; median 74. Picture Mushroom (24,664) leads, most of the rest have under 2,000, and “mushroom finder” (51, median 5), where several apps map where they grow, is even more open.
- Outfit finder. “outf”, 6th; 49; median 92. Find clothes from a photo: the biggest of the first 10 has 821 ratings.
- Wardrobe tracker. “wardr”, 4th; 41; median 431. Whering (12,024) and Stylebook (8,871) lead; the others have 0 to 1,647.
- Skincare routine. “skinc”, 4th; 41; median 447. SkinSort (6,460) leads, and nothing else passes 1,518.
- Tide times. “tide”, 10th; 41; median 0. Two big apps hold the top, Tides Near Me (163,620) and My Tide Times (17,597), then apps with 0 to 586 ratings.
- Horse racing games. “hors”, 10th; 41; median 51. Rival Stars (148,242) and Photo Finish (21,566) lead; the other racing and manager games have 0 to 6,395.
Searches that look like good ideas, and aren’t
The most searched phrases are the most tempting, and the most taken. Each of these is suggested after two or three letters:
None of these is impossible, but each means passing apps with tens or hundreds of thousands of ratings before reaching the first screen. Apple’s review guidelines say the same about some kinds of apps in particular:
Certain kinds of apps, such as dating, flashlight, sound effects, wallpaper, simple timers, and fortune telling, are well established on the App Store and we will not accept new submissions unless they offer a meaningfully different or improved experience.
The next sentence is blunter: “Other kinds of apps, such as drinking games, Kama Sutra, fart, and burp apps, are mediocre, low-quality, or low-effort and do not add value to the App Store.” Rule those out before anything else.
How were these app ideas found?
Each idea above is a search people make on the App Store, measured on 1 November 2026, in the United States, the way ASO Gauge measures a keyword:
- The words. We typed 85 everyday words into the App Store’s search box, such as “habit”, “dog”, “tip”, “guitar” or “ai plant”, and kept the 585 phrases of two to four words that it suggested.
- Demand. For each phrase, we found the fewest letters typed before the App Store suggested it, and its place then. ASO Gauge turns that into a popularity estimate from 5 to 95: 95, less 12 for each letter typed after the first and 2 for each place below the first, and 5 when the App Store never suggests the phrase.
- Competition. For the 353 phrases at 41 or more, we read the first 10 apps of the App Store app’s own iPhone search and took the median of their rating counts: half have fewer, half have more. Under 100 is what ASO Gauge calls low competition, under 1,000 medium, under 10,000 high, and 10,000 or more very high.
- Freshness. For the same apps, Apple’s public lookup gave the date of their last update and of their first release.
Ratings matter because Apple ranks search results on more than words:
Search results are based on a number of factors, including text relevance (matches for your app’s title, subtitle, keywords, and primary category), as well as user behavior (downloads, ratings and reviews, and more).
An app with 300,000 ratings has years of downloads behind it. A first 10 whose median app has 400 ratings is one a new app can join, if it answers the search better.
99 of the 353 searches had a median under 1,000. We then removed the names of existing apps, shops and places (people who type them want one thing, not a new app), the searches for medical tests and symptom checkers, two tools easy to misuse (a watermark remover and a receipt maker), and close variants of the same idea, kept as notes under it. 30 remain.
These are estimates, and they say so. The popularity comes from the App Store’s suggestions, not from Apple’s own search figures, which Apple shows only to Apple Ads advertisers. They describe the United States, on one night. A search that is open today can fill up next month: check the numbers again before you build.
How to check your own app idea
You don’t need a tool for the first test. You need the App Store and ten minutes.
- Write down the words a stranger would type to find your app, not your app’s name: three to five phrases of two or three words.
- Type each phrase into the App Store’s search box, one letter at a time. If it appears in the suggestions after three or four letters, people search for it. If it never appears, even typed whole, few people do.
- Search it, and look at the first 10 apps. Note their ratings and, on each product page, the date of the last update. Ten apps with hundreds of thousands of ratings mean years before you reach the first screen; ten with a few hundred, some not updated in years, mean room.
- Read their reviews. The one-star and two-star reviews of the top apps list what people wanted and didn’t get. That is your feature list.
- Check that the words fit your listing. Your app name and your subtitle have 30 characters each: the name and subtitle guide explains what Apple allows in them, and the keyword field holds 100 more characters.
- Make it different. Apple’s rule on copycats begins: “Come up with your own ideas. We know you have them, so make yours come to life.” Its rule on minimum functionality asks for “features, content, and UI that elevate it beyond a repackaged website.”
Google Play asks for the same in its policy: “Apps must be stable, functional, and provide meaningful content.” On both stores, a search with weak competition is a reason to build something good for it, not a reason to ship something thin.
Once the app is out, measure where it ranks for those searches, every day: the guide to measuring rankings explains how. ASO Gauge does it for you on the App Store and Google Play, and shows each keyword you track with the same popularity estimate as above, and on the App Store the same median of ratings.
Sources
- Apple, App Store, Search: how search results are ranked.
- Apple, App Review Guidelines: 1.4.1 on medical apps, 4.1 Copycats, 4.2 Minimum Functionality, 4.3 Spam.
- Google, Policy Center, Functionality, Content, and User Experience: what Google Play requires of an app.
- Measurements: the App Store app’s iPhone search and its search suggestions, United States, and Apple’s public lookup for each app’s last update, read the way ASO Gauge reads them, between 02:40 and 03:16, Paris time, on 1 November 2026: 585 searches from 85 words, 353 searched often, 99 with a median under 1,000 ratings.
Checked on 1 November 2026.