Method
How the S.I.F.T. score is calculated.
The score is a published, rules-based calculation from third-party data. It describes a company’s figures. It is not a forecast and it is not a recommendation to buy, sell, or hold.
Llsted is education and research software. You build your own screens and keep a dated journal of your own reasoning. The S.I.F.T. score is a published, rules-based calculation from third-party data; the method and weights are on this page and you can change the weights on your own book. Llsted never tells you what to buy, sell or hold, does not know your financial situation, and does not manage money or execute trades.
The four pillars
| Pillar | Inputs | Main thresholds |
|---|---|---|
| S — Safety | Moat classification, cash as a share of market value, net debt to operating cash flow, a conservative discounted-cash-flow value against the current price, the trend of return on invested capital (ROIC) | Cash ≥ 20% / ≥ 10% / < 5%; net debt to cash flow ≤ 2 / ≤ 5 / > 8 years; DCF gap ≥ 30% / ≥ 15% / < 0%; ROIC–WACC spread threshold of 10 points; ROIC trend over 3 years |
| I — Income stability | Years of continuous dividends, payout ratio, five-year free-cash-flow growth | Dividend years ≥ 25 / ≥ 10 / ≥ 5 / 0; payout 30–60% / 60–80% / > 80%; cash-flow growth > 10% / > 5% / < 0% |
| F — Fundamentals | Ten-year average return on equity, five-year free-cash-flow growth | ROE ≥ 20% / ≥ 15% / ≥ 10% / < 5%; cash-flow growth ≥ 15% / ≥ 8% / ≥ 0% |
| T — Time in market | Five-year beta, years to recover from the largest drawdown | Beta < 0.8 / < 1.0 / < 1.3 / ≥ 1.5; recovery ≤ 1 / ≤ 3 / > 5 years |
Weights and the final score
The house score weights each pillar equally (25% each). The pillars are combined with a weighted geometric mean, so one very weak pillar pulls the total down rather than hiding behind strong ones. The result is rounded to one decimal and kept between 0 and 100.
On your own book you can change the weights (for example through the onboarding quiz). Your weights change the number on your book only. They do not change the public page, and they are not a statement that a company suits you.
Bands
Strong (80 and above) · Adequate (60–79) · Caution (below 60). A band describes the score, not an action.
Missing data
A blank is a blank. When a figure is not available from the data provider it is shown as “Not available”. It is never filled in to make the page look finished. A company with no usable figures shows no score.
Model score (rules-based screen of the data shown; not advice or a forecast)
Company pages also show a percentage labelled “Model score (rules-based screen of the data shown; not advice or a forecast)”. It starts at 90% and is reduced by fixed rules; it never goes below 50%. Its inputs are:
- Coverage: minus 20 points if fewer than 5 years of continuous dividends are on file, the ten-year return on equity is missing, or cash as a share of market value is missing.
- Debt: minus 15 points if net debt is more than 8 years of operating cash flow.
- Price vs DCF: the same minus 15 points (applied once with debt) if the conservative discounted-cash-flow value is at or below the current price.
It is a rules-based screen of the data shown. It is not advice, not a forecast, and not a measure of how likely anything is to happen.
Data sources and updates
Figures come from third-party market-data providers. They can be delayed, wrong, or missing. Company pages are refreshed on a regular schedule; the date of the last refresh is shown on the page where available.
The role of AI
A language model writes a short plain-English summary of the pillar figures. It does not calculate or change the score, it is instructed never to recommend buying, selling, or holding, never to use a person’s name, and never to predict prices, and its output is filtered for those words before it is shown. It can still be wrong.
Version and change log
- Method v2.0 — four pillars, geometric mean, bands 80/60. This page first published October 2026.