Where the catalog comes from, and what it leaves out.

The dataset in numbers

Scroll horizontally to see all columns ↔

What each record contains

For every app: name, store category, first release date, latest version date, rating count, star average, listed download price and currency, developer name, store description, icon and store URL. Functions shown in the catalog are inferred from the description by keyword rules — they are a search aid, not developer-supplied ASO keywords.

The frozen Apple snapshot contains no review text, download count, revenue estimate, retention data, keyword volume or ranking history. Separately saved Google keyword reports are not additional fields supplied by Apple and do not refresh this snapshot.

Four biases to hold in mind

1. Recency is over-represented

Monthly first-release counts rise toward the capture date: 47,174 in September 2026 against 23,047 in March. This distribution may reflect removal and index coverage as well as launch activity. A single currently available listing snapshot cannot distinguish those effects or establish market growth. Inspect the monthly counts.

2. Delisted apps are invisible

An app that launched and was removed before capture does not appear at all. The catalog therefore overstates survival: the failures it shows are the ones that were still listed when we looked.

3. Established incumbents are excluded by design

Because the window is defined by first release, a market leader that shipped in 2019 is absent from every niche count. This is the most common way to misread the catalog: a niche can look leaderless and still have a dominant product.

4. Descriptions are marketing copy

The description field is written to sell. It is the best available signal about what a product does and the worst possible signal about whether it works.

What the catalog cannot answer

• How many people use an app. Rating counts are not downloads and are not comparable across apps.
• How much money an app makes. There is no revenue field; listed price covers downloads only, not subscriptions.
• Whether users are satisfied. Star averages cluster near 4.5 on tiny samples and there is no review text.
• What people search for. Apple snapshot fields contain no Google search-volume or traffic data. The catalog can show separately saved keyword reports; read each report's date, market and provider estimates rather than treating them as app usage or demand.
• What is available today. The snapshot is frozen at 23 Oct 2026; apps may have been removed or repriced since.

How often it refreshes

The catalog is refreshed as a new immutable snapshot, not patched in place. When a refresh lands, pages that cite its counts must identify which facts were rechecked and record the updated analysis scope. Until then, the published Apple snapshot counts describe 23 October 2026. Separate keyword-report dates and text-correction dates do not refresh those counts.

Verify the snapshot and research rule

Public evidence can be read and downloaded without an account or payment. The snapshot verification summary records the exact UTC window, collection totals, category and monthly counts. These describe this capture, not all apps ever released.

The invoicing review uses rule estimates-summary-v1: Business, Productivity or Finance; a case-insensitive OR substring search for invoice, estimate, quote or billing in the stored first 180 characters of the description. Full descriptions are read only after this screening step.

• Candidate manifest (JSON): 1,474 IDs, matching summaries, captured facts, the rule and a SHA-256 digest of its records.
• Candidate facts (CSV): the same public research set in a spreadsheet-friendly form, without full descriptions or artwork.
• Rating buckets: 961 with zero, 497 with 1–49, and 16 with 50+. A 50-rating threshold is an editorial reading choice.

The manifest lets a reader check the published set’s membership and arithmetic. It does not by itself prove completeness against the original capture. A local verification command, npm run research:verify, compares every published candidate fact against the original read-only catalog and rechecks full-description mention counts. It needs the original snapshot; normal site builds use the checked-in research input and do not recollect data.

The distinction matters: applying the same substring terms to full descriptions returns 7,089 candidates and 82 with 50+ ratings. That broader set has not received the review published here. See the scope correction.

Use and attribution

The underlying data comes from Apple's public sitemap and lookup endpoints and describes third-party apps. The catalog is an independent index and is not affiliated with, endorsed by, or sponsored by Apple. App names, icons and store descriptions belong to their developers. If you reuse the counts published here, cite the snapshot date — without it the numbers are meaningless.

What this site republishes, and what it does not

• Facts — name, developer, category, first release date, listed price, rating count and star average — are recorded as facts about a public listing.
• Descriptions are excerpts. The catalog stores a 180-character excerpt for search, and the detail view publishes at most 700 characters followed by a link to the developer's own App Store page, which is the authoritative and current copy. The full text is not redistributed.
• Icons are referenced, not copied. Where an icon appears it is loaded from Apple's image CDN and links to the store listing; the artwork itself belongs to the developer.
• Nothing here is an opinion from Apple, and no listing placement, ranking or removal has been paid for.

If you are a developer and want a listing corrected or removed, write to hello@saasnichefinder.com with the app name and the change; removals are honoured without argument and the change is noted on this page.