App Store Trends & AI App-Idea Report

Find out what's rising on the Apple App Store in any country and category, what users complain about in the apps at the top, and get an AI-written report of concrete app ideas based on those unmet needs. Built for indie developers, product managers, app studios and investors who want ideas grounded in real chart and review data, not guesswork.

Data source. Apple's public App Store RSS feeds (charts and customer reviews) and the iTunes Lookup API. No scraping of web pages, no login. This Actor is independent and not affiliated with, endorsed or sponsored by Apple Inc. App Store is a trademark of Apple Inc. Chart positions are Apple's own; this Actor doesn't create its own rankings.

What does it do?

1. Reads an App Store chart (top free, top paid, top grossing, new free or new paid; up to 100 positions) for any of Apple's country storefronts, overall or for one category (Productivity, Health & Fitness, Finance, Games...).
2. Tracks movement between runs. Schedule it daily or weekly and each chart row shows previousRank, rankChange and isNewEntry. A simple, explainable momentumScore favours apps that climbed, are new to the chart, or were released in the last 90 days.
3. Analyses the most interesting apps. It reads their most recent reviews (up to 500 per app) and returns a recent average, star distribution, share of 1–2 star reviews, review rate, the most common complaint and praise terms, and a few short, anonymised review excerpts.
4. Writes an app-idea report (optional) in Markdown. An AI model writes the summary and 5–6 concrete app ideas, each with target user, the review evidence behind it, must-have features, how it could make money and the main risk. The "what users complain about", "what's rising" and caveats sections are computed directly from the data, so their numbers and quotes are real. The AI part is clearly labelled, and any quote the AI uses that doesn't match a collected review is flagged.

Use cases

• Indie developers: find an underserved niche before you build.
• Product teams: watch competitors' ratings and complaint themes after each release.
• App studios and agencies: weekly market briefings per category and country.
• Investors and analysts: spot fast climbers and new entrants early.

Input example

{
  "country": "us",
  "chart": "top-free",
  "genre": "productivity",
  "appsToAnalyse": 8,
  "pickAppsBy": "rising",
  "reviewsPerApp": 100,
  "generateReport": true,
  "reportFocus": "ideas a solo developer could build in a month"
}

You can also add appIds (App Store IDs or app links) to analyse your own app or specific competitors alongside the chart.

Output

• Dataset, view "Chart & momentum": one row per chart position, with rank movement, all-time rating and rating count, price and release date.
• Dataset, view "App analysis": one row per analysed app, with review statistics, complaint and praise terms, and anonymised excerpts.
• Key-value store record REPORT: the AI report as Markdown (also in the dataset as a report row).
• SUMMARY record: counts, timing and any reason the report was skipped.

Example app-analysis row (shortened, from a real run on the NZ Productivity chart on 27 Oct 2026):

{
  "type": "app",
  "name": "Microsoft Authenticator",
  "rank": 3,
  "reviewsAnalysed": 92,
  "recentAverageRating": 1.93,
  "negativeShare": 0.739,
  "complaintTerms": [{"term": "phone", "reviews": 22}, {"term": "account", "reviews": 13}, {"term": "code", "reviews": 11}],
  "complaintExcerpts": [{"rating": 1, "title": "…", "text": "…", "version": "6.8", "date": "2026-10-20T…"}]
}

Sample report (real run, 28 Oct 2026 NZDT)

Input: {"country": "us", "chart": "top-free", "genre": "health-fitness", "chartLimit": 50, "appsToAnalyse": 5, "reportFocus": "ideas a solo developer could build and launch in about a month"}. Took 26 seconds. Shortened excerpt of the REPORT record:

What users complain about (computed from the reviews, not written by AI)

• Planet Fitness (#3, recent avg 1.72★ from 100 reviews, 79% negative): update (24), check (24), new (21), back (12), membership (11).
• Flo Cycle & Period Tracker (#2, recent avg 2.73★ from 100 reviews, 54% negative): pay (18), free (10), money (9), subscription (8), tracking (8).
• Strava: Run, Bike, Walk (#1, recent avg 3.1★ from 100 reviews, 43% negative): subscription (8), update (8), new (7), annoying (6).

App idea 1 (AI-written): "Cycle Sync Companion": A free, privacy-focused period tracker that offers core tracking without a subscription.

• Evidence: Flo Cycle & Period Tracker: pay (18 reviews), free (10 reviews), money (9 reviews), subscription (8 reviews), deleted (7 reviews), paid (7 reviews), information (6 reviews).
• Must-haves: Accurate cycle prediction, symptom logging, data export, no subscription for core features.
• Risk: Competing with established, feature-rich (but often paid) apps.

The full report also has a summary, 4 more ideas, a "What's rising" list and caveats.

Pricing (pay per event)

Event	Price
Actor start	$0.005 per run
Chart app (one chart row)	$0.001
App analysed (reviews + stats)	$0.02 per app
AI report	$0.50 per report

AI model usage is billed separately. The report is written by an AI model through Apify's own OpenRouter proxy (apify/openrouter), which charges your Apify account the model's cost (with the default google/gemini-2.5-flash, about $0.01 per report on paid Apify plans in our tests; Apify charges free-plan accounts about 10 times more, roughly $0.08). A typical run (100 chart rows, 10 analysed apps, one report) costs about $0.81 plus the AI usage. A smaller run (50 chart rows, 5 analysed apps, one report) costs about $0.66. Set a maximum cost per run and the Actor stops cleanly before going over it. If the report can't be generated, you aren't charged for it.

Good to know

• Reviews: Apple's feed returns at most the 500 most recent reviews per app and country, so statistics describe recent sentiment, not all-time ratings (use averageRating and ratingCount for those).
• Rank changes need a previous run for the same country, chart and category. Your first run shows rankChange: null; schedule the Actor to build history.
• No downloads or revenue data: Apple doesn't publish them, and the report says so.
• Privacy: reviewer names and profile links are never collected; emails, links and phone numbers are removed from review text.
• AI report: the AI part may contain mistakes. It only gets the data collected in the run and is told not to invent numbers; the dataset report row lists any unverifiedQuotes and whether the AI output was truncated.

FAQ

Which countries are supported? Any App Store storefront, by its 2-letter code (us, gb, de, jp, nz, ...).

Can I track my own app? Yes: add it to appIds and schedule the Actor to see rating and complaint trends after each release.

Can I use a different AI model? Yes: set aiModel to any model ID offered through Apify's OpenRouter proxy, for example openai/gpt-4o-mini (cheaper, but its ideas were noticeably more generic in our tests). Output is capped at about 2,000 tokens by the proxy; if a model runs over, the report says it was cut off.

Is it legal? It uses Apple's public RSS feeds and Lookup API, keeps excerpts short and anonymised, and doesn't republish full review text or Apple artwork. Check that your own use follows Apple's terms and the laws that apply to you.