App Store New Apps Scraper: iOS Publisher Leads

For app-growth agencies and investors who want newly released iPhone apps, their dated first releases, publishers and source links without collecting app IDs first. About $0.0026 per verified new app, or $0.26 for 100 apps, plus the minimum $0.00005 Actor start charge. No Apple account or API key is needed.

Each row gives the app, publisher, original release date and source URLs. Websites appear only when Apple supplies them. Coverage is limited to public rolling feeds, not every App Store launch.

What counts as new

The Actor reads Apple's newapplications feed for each selected storefront. It parses the original release timestamp, excludes dates outside your requested window, removes duplicate app IDs and checks the remaining apps through Apple's public lookup endpoint. The lookup's releaseDate must agree with the feed. A recent currentVersionReleaseDate does not make an older app a new launch.

dateVerified: true means the original date was confirmed against Apple's lookup data. It does not mean an independent investigation established when the business opened, when development began or when the product first launched anywhere in the world. Dates can differ between storefronts. Conflicting feed and lookup dates are excluded without an app charge.

Publishers can be individuals or companies. publisherName is Apple's artist/publisher label, and publisherType remains unknown. The Actor does not infer corporate status from a business-sounding name or identify founders. There is no verified-company-only filter, email extraction or guessed domain enrichment.

What you get

Field	Meaning
verdict	Readable lead summary: app, publisher, dated release, stores, price and website availability.
appId	Stable App Store app ID, stored as text. One paid row per distinct ID.
appName	App name returned by the country-specific lookup.
publisherName	Apple's publisher/artist name; not a verified company identity.
publisherId	Apple's publisher ID when supplied, otherwise null.
publisherType	Always unknown; no corporate verification is performed.
originalReleaseDate	Original release timestamp in UTC, confirmed against lookup data.
currentVersionReleaseDate	Latest version timestamp when present; never used as the first-release date.
observedAt	When the verified record was collected. This is not a launch date.
availableCountries	Queried feed storefronts containing the qualifying app. Not worldwide availability.
category	Primary Apple category, such as Productivity or Shopping.
categoryId	Primary Apple category ID as text.
price	Numeric price in the lookup storefront, or null when absent. Zero means free.
currency	Supplied currency code, or null.
priceCountry	Storefront used for the returned price and lookup details.
appStoreUrl	Link to the app's App Store page.
publisherWebsite	Apple's supplied sellerUrl, or null. It is not guessed or independently checked.
sourceUrls	Public feed and individual lookup URLs for inspection.
dateVerified	True for delivered app rows that passed original-date validation.
coverageNote	Explanation of the evidence and limits for this record.

The default dataset contains app rows. Its publisher view puts publisher identity first for review. The free PUBLISHERS output groups those same apps into one record per publisher, and OUTPUT links to a separate grouped dataset. Grouping is free and reduces repetitive outreach.

Every run also saves SUMMARY: queried countries, fixed date window, stale and failed feeds, old/future/duplicate exclusions, lookup exclusions, output count and spending-limit status. Empty runs still have a summary. A correct zero-result response means no qualifying records were available for your filters; it does not prove no apps launched.

Start in three steps

1. Select storefronts and a first-release window. The default is us and 7 days.
2. Set maxApps. The form's five-app prefill is a quick first check; the runtime default is 25.
3. Run the Actor, inspect the summary and export the app dataset or publisher groups.

The initial five-app US cloud test on 15 October 2026 completed in 7.4 seconds. It read 110 feed entries, excluded four future-dated entries and delivered exactly five verified apps. Apple's feed URL includes limit=10, but the actual feed ignored that as a strict cap. The Actor enforces its own output limit.

Inputs and examples

Input	Default	How to use it
countries	["us"]	List of tested two-letter storefront codes. Duplicate codes are removed.
since	"7 days"	Date, timezone-aware timestamp, today, yesterday, or a duration such as 2 weeks.
maxApps	25	Maximum distinct delivered apps, from 1 to 1,000. Prefill is 5.
keywords	none	Case-insensitive OR substring matches in app name, publisher or category.
categories	none	OR matches against exact primary category names or numeric IDs.
excludeGames	false	Exclude the Games category and apps whose lookup genres include Games.
publisherTypes	"any"	Only any is supported; publisher type remains unknown.
skipPreviouslyDelivered	false	Enable persistent app-ID deduplication across sequential runs of one watch.
watchId	"default"	Stable delivery-history name. Use separate names for separate prospecting lists.
appIds	none	Optional known-ID verification instead of feed discovery, using the first country.

Relative durations start at midnight UTC that many days before the run. A plain date also starts at midnight UTC. The upper bound is the run's initial timestamp. Both bounds remain fixed if the run restarts. A wider window does not retrieve archived feed entries that Apple no longer exposes.

Keywords and categories filter the available feed; they do not search the complete App Store catalogue. For example, keywords: ["fitness", "training"] keeps a record matching either word. Combining that with a category filter requires both filter groups to match. A narrow filter can legitimately produce no rows.

{
  "countries": ["us", "gb", "ca"],
  "since": "7 days",
  "maxApps": 100,
  "excludeGames": true
}

For a daily prospecting task, use a stable watch name and schedule sequential runs:

{
  "countries": ["us"],
  "since": "7 days",
  "maxApps": 100,
  "skipPreviouslyDelivered": true,
  "watchId": "agency-us"
}

Run each watch sequentially. An unfinished previous run produces a free watch-busy summary; resume that original run before reusing the watch. Separate watch IDs keep separate histories; changing the ID can return and charge previously delivered apps again.

Storefront coverage

The tested codes are us, gb, ca, au, de, fr, es, it, nl, se, no, dk, fi, ie, nz, za, in, jp, kr and br.

In the 15 October 2026 source test, 13 of 20 feeds contained qualifying records for the October 8-15 window. All 20 returned structured data. Seven were stale for that window. A feed can respond normally while listing old apps, so HTTP success alone does not establish freshness. Fresh storefronts were US, GB, CA, DE, FR, NL, SE, DK, FI, IE, ZA, IN and BR. Stale storefronts were AU, ES, IT, NO, NZ, JP and KR. Check SUMMARY for current coverage.

Real output example

US prefill example from 15 October 2026; the dataset includes additional audit fields.

{
  "verdict": "New Shopping app StylerOutlet- Gift Ideas Shop by MageComp, first released 15 Oct 2026 (US store), free, no website listed.",
  "appId": "6809700152",
  "originalReleaseDate": "2026-10-15T12:33:26.000Z",
  "appName": "StylerOutlet- Gift Ideas Shop",
  "publisherName": "MageComp",
  "publisherType": "unknown",
  "category": "Shopping",
  "price": 0,
  "currency": "USD",
  "priceCountry": "us",
  "publisherWebsite": null,
  "availableCountries": ["us"],
  "dateVerified": true
}

Pricing and spending limits

You pay once per distinct verified app delivered, plus Apify's minimum start charge. No extra default dataset-row fee is added. Publisher groups, summaries, duplicates, old/future exclusions, missing app IDs and failed verifications have no app charge.

Delivered apps	Base app charge	Bronze app charge
5	$0.013	$0.01235
25	$0.065	$0.06175
100	$0.26	$0.247
1,000	$2.60	$2.47

Add the $0.00005 minimum start charge to those examples. Silver is $0.00234 per app and Gold is $0.00208 per app. Base pricing is $0.0026; Bronze is $0.00247. The Actor has no additional outside API fees.

Set Apify's maximum charge to control spending. The Actor checks remaining capacity before enrichment and delivery, stops cleanly at the limit and records that decision in the summary. A low spending limit can reduce output below maxApps. An empty run incurs the platform start charge but no app charge.

API and automation

Use the normal Apify API with your own token. Keep API tokens private.

curl -X POST \
  'https://api.apify.com/v2/acts/moggedhq~app-store-new-apps/run-sync-get-dataset-items' \
  -H 'Authorization: Bearer YOUR_APIFY_TOKEN' \
  -H 'Content-Type: application/json' \
  -d '{"countries":["us"],"since":"7 days","maxApps":5}'

from apify_client import ApifyClient
client = ApifyClient("YOUR_APIFY_TOKEN")
run = client.actor("moggedhq/app-store-new-apps").call(
    run_input={"countries": ["us"], "since": "7 days", "maxApps": 25}
)
rows = list(client.dataset(run["defaultDatasetId"]).iterate_items())

import { ApifyClient } from 'apify-client';
const client = new ApifyClient({ token: process.env.APIFY_TOKEN });
const run = await client.actor('moggedhq/app-store-new-apps').call({
  countries: ['us'], since: '7 days', maxApps: 25
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();

Save a task, then connect its completed-run webhook to Make, Zapier, n8n or Sheets. For larger jobs, start asynchronously and fetch SUMMARY alongside the completed dataset.

Use with AI agents (MCP)

Connect an agent to https://mcp.apify.com and ask it to find App Store new apps with first-release dates, publishers and supplied websites. For a direct run, name moggedhq/app-store-new-apps and specify countries, since and maxApps. Ask the agent to inspect SUMMARY as well as the dataset, preserve null websites and avoid presenting publisher labels as verified companies. Discovery needs no supplied IDs.

Reliability and limits

Requests are sequential within each run, roughly three seconds apart. Retries use fresh proxy sessions and honour Retry-After. Datacenter proxies are tried first, with residential only as the final retry. Source responses are byte-bounded, and feed processing is capped independently of Apple's ignored limit parameter.

Verified rows, the original window, country state and candidate snapshots survive a reboot. Delivery and charging are serialized. Saved rows are reconciled with authoritative run event counts and durable acknowledged-charge receipts, so recovery does not rely only on a temporary charge key. If that evidence is ambiguous, the Actor stops rather than risking another charge. This protects delivery; it cannot make an unavailable or stale source complete.

FAQ

Does this find every newly released iPhone app? No. It finds qualifying records in the queried public rolling feeds. Apps absent from those feeds are outside coverage.

Can I backfill the last year? A wider since date filters the current snapshot only. There is no complete historical backfill.

Is a version update counted as a new launch? No. Qualification uses original releaseDate, never the current version date or observation time.

Are publishers verified businesses? No. Publisher names may represent people or companies. Every publisherType remains unknown.

Why is a website missing? Apple did not supply a usable sellerUrl. No domain is inferred and no website crawling is performed.

Why did a country return zero apps? Its feed may be stale, your filters may have no matches, or no records may pass date verification. SUMMARY distinguishes these cases from failed source reads.

Will the same app cost more across several countries? One run deduplicates app IDs and charges once. Enable a stable watch to skip apps delivered on earlier sequential runs.

Can I enrich known IDs? Yes. Supply appIds; the first country determines lookup details. The same original-date window and filters still apply, so an old app can be excluded even when its metadata is available.

Are grouped publishers extra? No. The publisher dataset is a free grouping of the app rows already delivered.

What happens with invalid input? Inputs the Actor can parse become a free explanatory note and summary. Apify may reject schema-invalid requests before starting; no app is charged in either case.