Store intelligencePrice, plans, privacy disclosures, package contents, and public-review patterns.
Listed
Independent app intelligence
The missing trust layer for the app stores.
App stores are excellent at showing what is popular. Appworthy adds the part people still have to discover the hard way: the real price, the pressure tactics, the cancellation path, and the evidence behind every claim.
HOW A LISTING GAINS AUTHORITYAPPWORTHY
Automated evidenceReproducible facts and independent signals are fused into a row of published measurements.
MeasuredHuman reviewA person runs our test protocol on a real device and records what they measure — confirming or contradicting findings, and adding measurements no artefact can supply.
ReviewedPayment funds the work. It never buys a result, placement, or a favourable finding.
Our operating principles
Trust comes from visible constraints.
The product is designed so the strongest-looking claim also needs the strongest evidence.
Show the catch
Card-gated trials, weekly billing, hard paywalls, forced accounts, cancellation friction, and aggressive advertising belong on the listing—not in the fine print.
Separate money from judgment
If hands-on tests are ever sold, a developer pays for the labour. Nobody can pay for a favourable measurement, a mark, or placement above a fairer app.
Let evidence set the claim
A reproducible fact can stand on its own. An interpretation needs independent source kinds to agree. A lone signal stays visible, labelled “One source only”, and off the catch list.
From listing to finding
How Appworthy analyzes an app
Automation builds breadth. Corroboration controls what reaches the catch list. A hands-on test adds what only a person with the app in hand can measure.
Collect what the stores disclose
We capture prices, in-app purchase ladders, privacy labels, release history, ratings, and the developer's own listing claims.
Inspect the app itself
Package scans look for SDKs, permissions, billing components, trackers, and the real paywall or cancellation language shipped inside the app.
Find recurring user evidence
Public reviews are grouped into pain-point themes. Repeated reports count as one source kind, so volume cannot manufacture independence.
Fuse and publish the evidence
Measured facts may publish directly. Interpretations reach the catch list only when independent sources agree — or, for a harm only users can see, when enough reviewers clear that charge's own published bar. Every report says what each finding rests on.
One meaning
What we measure
Consumer financial fairness: what an app does with your money, attention, and consent. It is not a quality rating or a popularity contest.
WHAT WE PUBLISH
Bands
Eight measurements per app, each published as a count rather than a verdict. Five of them need no package scan and reach the whole analysed catalogue; three more are added wherever we have downloaded and read the app itself.
?“Not assessed” is its own answer, and it is the common one
Pricing & trial transparencyReal cost, renewal terms, pricing ladders
Free-tier honestyPaywall timing, usable free product, pay-to-win
Exit & refund fairnessCancellation, billing stops, refund paths
Pressure & dark patternsUrgency, preselection, misleading disclosures
Ads & attentionAdvertising volume, placement, and ad stack
The five headings above are how findings are organised. They do not add up to a published number. A 0–99 score was published here until the graded bands replaced it: measured over the catalogue it left 44.7% of every pair of apps tied, so it could not tell most apps apart — and, more importantly, a single number is a verdict where a count is an observation. The full reasoning is on the methodology page.
A hands-on test measures what automation cannot: a person runs our test protocol on a real device, and the test confirms or contradicts findings and adds measurements. It does not change a band. Three commitments come with it: every completed test is listed on the audit record with what it confirmed, added and contradicted, so nobody has to take our word for it; a human-reviewed app carries a distinct dated mark, never folded into a band row; and we run tests at our own cost, so the mark is never reachable only by paying. No human review has been completed yet: every measurement on this site today is automated.
Claim strength
Not every signal gets the same authority.
The label on a finding tells you exactly how much evidence is behind it—and whether it is on the catch list.
Store facts and a band row
Every listing carries its band row. A line we could not measure says “Not assessed” and gives the reason. It is never shown as a zero.
One source noticed it
The signal is visible so readers can inspect it, but it is explicitly not something we would stand a measurement on.
Independent evidence agrees
Two source kinds put an interpretation on the catch list as “Corroborated”; three or more as “Strongly corroborated”.
Hands-on test
A person runs our test protocol on a real device and records what they measure, leaving a dated “Human-reviewed” mark that has no end date. Every completed test is listed on the audit record.
Two inputs are required before any automated measurement appears.We need enough public-review analysis and at least one app-side source that actually observed material. Failure to inspect something is never treated as proof that it is clean.
For iOS & Android developers
Inspection programme — not yet open
Appworthy will offer a hands-on test: a person runs our test protocol against your app on a real device and records what it measures. The programme has not opened. Nothing is on sale, and no price is set.
- Bands are free and automatic for every analysed app. There is no band you can only get by paying.
- What you would buy is the test, not the outcome. A test can go against you as readily as for you, and it does not change a band.
- The audit record is published: every completed test, with what it confirmed, added and contradicted.
Factual corrections are free and always will be. If something objective on your listing is wrong, send the evidence and we correct it.
Choose apps with the evidence in view.
Browse the catalog, inspect the catch, and open the report behind every assessment.