How KvantraRank works
The three factors
Value
40%
How cheaply a company is priced relative to its fundamentals. Cheaper companies have historically earned higher returns over time.
Quality
35%
How solid and well-run a company is: profitability, cash generation and balance-sheet strength.
Momentum
25%
Whether the company has outperformed its peers over the past year, including the trend in analyst expectations.
These three factors are chosen because they are the best documented in academic research across markets, time periods and sectors, which is why the Norwegian sovereign wealth fund uses the same framework.
The weighting
KvantraRank starts from a base weighting of Value 40%, Quality 35% and Momentum 25%. Combined, these identify attractively priced quality companies with positive price trends.
These weights adapt slowly to the market. When valuations within a region and sector are unusually stretched (a wide gap between the cheapest and most expensive companies) the Value weight is nudged up (and Quality and Momentum down) because cheapness tends to matter more in those conditions. The opposite happens when the spread is narrow. The adjustment is capped at ±10 percentage points and is measured against years of history, so the weights drift gradually rather than jumping around.
Fair comparison
A stock is only ranked against other companies in the same region and sector, never against the entire market. A Norwegian industrial company is compared with other European industrials, not with US technology companies. We also separate large caps from small- and mid-caps, so companies compete against real peers.
Tested against history
Research says Quality, Value and Momentum have paid off over time. We wanted to prove that KvantraRank does too. From January 2000 to September 2026, at the start of each month, we bought every stock with a KvantraRank of 80 or higher, held them for a month, then sorted again. Before costs, the group returned 28.66% a year. That is almost 27 years, through the dot-com crash, the 2008 financial crisis and the COVID crash.
| Line | Annualized return |
|---|---|
| KvantraRank 80–100 | 28.66% |
| Nasdaq-100 (QQQ) | 8.68% |
| S&P 500 (SPY) | 8.28% |
The S&P 500 line is the SPY index fund with dividends reinvested, in US dollars. The Nasdaq-100 line is the QQQ index fund with dividends reinvested, in US dollars.
Simulated results. This is a backtest: the returns were calculated after the fact using historical data and were not earned by a real portfolio. It does not include trading costs, taxes or fees. Past returns, real or simulated, say nothing about future returns. This is not investment advice.
Which stocks. Every company our scores cover is in the test once, on its main listing. There is no minimum company size or trading volume. A stock is left out if it hasn’t traded in the 10 days before the month starts, or can’t be bought within the first 5 days of the month. Stocks that trade rarely can still be included, and would be harder to buy in practice. Companies that later left the market are included when our data still has them: 12,073 of the 16,355 that were delisted between 2000 and 2026.
How the test worked. At the start of each month, every stock got a KvantraRank, using only information that was public at the time. The score works the same way as in Kvantra today: it looks at Value, Quality and Momentum, and compares each stock with similar companies in the same sector, region and size. We then bought all stocks with a score of 80 or higher, in equal amounts, and held them for one month. Then we did it all again. Dividends were reinvested, and a company that went bankrupt counts as a total loss.
The details. Financial statements count from the day they were filed. Where the filing date is missing or unreliable, they count from three months after the period ended, so a company that reported later than that has its figures used slightly too early. The test differs from the live score in two ways. Our history of analyst data is incomplete, so Momentum in the test is based on price only, and the Value metrics that use analyst estimates for the next 12 months uses the last 12 months of reported numbers instead. The test therefore scores more stocks than the live KvantraRank does. Stocks were bought at the first traded price of the month, and dividends were reinvested before withholding tax. Each stock’s return is calculated in its own currency. With all returns converted to US dollars instead, the result was 27.95% a year.
What the test does not include. No brokerage fees, taxes, or the small loss you take on every trade from the gap between buying and selling prices. The group replaces much of its holdings every month, so these costs add up. A real investor following the same rules would have earned less than the chart shows.
Compared with what. The green line shows what you would have got by buying every stock with a KvantraRank of 80 or higher. The chart also shows the S&P 500 and the Nasdaq-100, each a fund with dividends reinvested, so you can see how those markets did over the same period. Each gives bigger companies more weight, so they work a little differently from our test.
Where you'll see it
You'll find KvantraRank everywhere on the platform. The KvantraRank table lets you rank every stock in a country. The screener lets you filter stocks by score. Each stock profile shows its score, and a strip under the price chart shows how the score has moved over time. Your portfolio shows the combined score of your holdings.
When data is missing
Some companies don't have enough data for a reliable score. A sub-score (Quality, Value or Momentum) is only approved when enough of its underlying metrics are available. The combined KvantraRank is only shown when all three sub-scores are approved, so a KvantraRank always means the same thing: a complete Quality, Value and Momentum assessment. If one sub-score is missing, we show the two sub-scores we have and leave the KvantraRank as N/A. If too many metrics are unavailable, no scores are shown at all. We never re-weight a score on partial data.
Academic grounding
The approach builds on decades of factor research, including Novy-Marx (2012) on gross profitability, Sloan (1996) on accruals, and Cooper, Gulen & Schill (2008) on asset growth, among others.
Rankings are based on a mechanical quantitative model and do not constitute a recommendation to any specific person. Kvantra provides data and quantitative analysis only. Nothing here is investment advice. Historical returns do not guarantee future returns. All investments carry risk.
- Vidar Aasestrand“The dashboard is very clear and easy to use”
- Mathias“Quick and easy to create an account”
- Alex Hagerup“Helped me invest better”
- Øyvind L.“a professional and user-friendly platform for stock analysis”
- Thias_ Brunvoll“helps me make good decisions”
- Tor Marthin“Clear layout and good visualization”
- S.W.R“The KvantraRank feature is super helpful”
