Data Sources & Reliability
Our valuations are only as good as the data behind them. This page explains transparently what data we use, how fresh it is, the quality checks we run, and where third-party and company-reported data have their limits.
Prices refresh every trading day, fundamentals with each new filing. Automated plausibility checks catch outliers, currency and share-class errors. When in doubt we would rather withhold a valuation than show a wrong one.
Where we stand, in numbers
- 40,000+ stocks valued worldwide, each with up to 26 valuation models and up to 37 quality factors.
- Every stock carries an evidence status. Where the data is too thin we show no fair value rather than guess.
- Automated checks run every day over prices, units, currencies and the consistency of the numbers across every surface.
- Online and on public record since 20 March 2016: timeline with sources.
What data we use
- Fundamentals from filings, income statement, balance sheet, cash-flow, multi-year history.
- Market prices, end-of-day closes from each home exchange, currency-normalised.
- Ratios, margins, growth, returns on capital, leverage, dividends, derived from the raw figures.
- Market data, market capitalisation, share count, sector/industry, country, currency.
Freshness
Prices refresh every trading day after the close. Fundamentals change as soon as a company reports new figures. Every stock page shows the date of the underlying fair value so you can see how fresh the read is.
Quality checks
Before a valuation becomes visible it passes automated plausibility checks that, among others, catch:
- implausible outliers (e.g. extreme fair-value jumps) are withheld rather than published;
- currency mislabels (ADRs whose local currency is reported as USD) are caught via a market-cap cross-check and corrected or withheld;
- share-class and dual-listing distortions are reconciled via market cap ÷ price;
- an ongoing cross-validation checks that the numbers shown stay consistent across every surface.
How we use AI
AI helps us interpret, not calculate. The fair value comes from reported figures and fixed models. AI rates soft factors such as brand and competitive advantage, sums up a short verdict with the main risk per stock, and cross-checks data. The rule: our data provider's figures come first. Only where a figure is missing or provably wrong may AI research enter the figure from the company's own annual report, and only when two independent checks find the same value. If the provider later supplies a consistent figure, that figure applies again. Such stocks carry the note "Data corrected by AI research".
Limits of third-party data
Fundamentals come from company disclosures and data vendors. They can contain errors, be restated later, or arrive late. One-off items, differing accounting standards and exchange rates can distort individual figures. We screen and filter automatically, but cannot guarantee every single reported number. When in doubt we would rather withhold a valuation than show a wrong one.
How to interpret the output
Always read a stock page together: fair value and evidence status and quality score and the bear/bull range. A large undervaluation with low evidence or weak quality is a flag for deeper research, not a signal on its own. There is no guarantee of returns, and nothing here is financial advice.
What some of our 4,480 users say
-
Trustpilot ★★★★★
“extensive data in format that is easy to understand”
Martin Grovdahl, USAOct 2026
-
Trustpilot · translated ★★★★★
“I appreciate the good overview and the wealth of metrics when you want to look at a stock more closely. The fair value estimate is sometimes more cautious than elsewhere, which I see as a positive. There is even personal contact: a human replies, not a bot. Very satisfied, and I recommend this portal.”
Andreas J., SwitzerlandOct 2026
-
Trustpilot ★★★★★
“This tool is great. I use it every time I want to buy a stock or study a stock. You get a valid calculation if you're buying an over- or undervalued stock. A perfect tool.”
Samad Golzari, GermanyAug 2026
-
Email · translated
“Impressive how quickly you implemented that. Thank you, that really helps me.”
DanielAug 2026
-
Instagram
“That was a super quick fix, Peter. Amazing. Your application is a gem for a beginner like me.”
Hemant P., IndiaAug 2026
-
Forum
“The interface is really well done, despite all the information it holds.”
MatthesAug 2026
-
Forum
“Cool tool, I hadn't come across it before. Impressive.”
Lever H.Aug 2026
-
Forum
“Even with a great deal of information available, it comes across as quite clear and easy to understand.”
hattifnattJul 2026
-
Trustpilot ★★★★★
“Amazing Tool. Quick & easy. Exactly what I was looking for! Chapeau! I would recommend it to everyone!”
Gunther Hingsammer, AustriaJul 2026
-
Pro user
“Not just an amazing website, but also amazing customer service. The error in my watchlist is fixed. Thank you very much for quickly fixing the problem.”
verified Pro userJul 2026
Frequently asked questions
Where does the data come from?
Fundamentals from company filings (income statement, balance sheet, cash flow, multi-year history) and end-of-day closes from each home exchange, both from a professional financial-data provider. Ratios such as margins, growth and leverage we compute ourselves from the raw figures.
How current are prices and fair values?
Prices refresh every trading day after the close, fundamentals as soon as a company reports new figures. Every stock page shows the date of the underlying fair value.
What happens when the data is wrong?
Before a valuation becomes visible, automated checks look for outliers, units, currencies and share classes. When in doubt we withhold a valuation and say so, rather than show a wrong number.
Does an AI calculate the fair value?
No. The fair value comes from reported figures and fixed models. AI rates soft factors such as brand and competitive advantage and cross-checks data. Our data provider's figures come first. If a figure is missing or provably wrong, AI research may enter the figure from the company's own annual report, but only when two independent checks find the same value. Such stocks carry the note "Data corrected by AI research".
Who stands behind it

Dr. Peter Klein
Founder, Graz, Austria. Dentist, and a private investor since 2004.
I built the calculator for myself first, because I wanted to know what a stock is worth before I buy it. It has been public since March 2016, and I have worked on it almost every day since.