The dividend stock universe

Last updated 12 September 2026.

The dividend stock screener covers a curated list of companies rather than everything that trades. This page says what is on that list, what is not, and how each figure in the table is worked out.

Everything here is informational. It is not financial advice, not a prediction of future returns, and not a recommendation to buy or sell.

What is in it

Companies that pay a dividend, listed in London or on the US exchanges, above a market capitalisation floor.

The dividend requirement comes first. A company that pays nothing has nothing to screen on here, so it is not in the universe at all. That is a stricter rule than it sounds: a company that suspends its dividend eventually leaves.

The cap floors are in each listing's own currency. A London listing is measured in pounds and a US listing in dollars, because that is how the figures arrive and converting them to a single currency to apply one threshold would quietly change which companies qualify as exchange rates moved. It does mean the UK and US floors are not identical in real terms, and that is the honest trade rather than a rounding error.

The UK is covered more deeply than the US. Sharesight and Snowball do not do UK depth well and Simply Safe Dividends does not cover the UK at all, so being thorough about London is worth more than matching US breadth.

The list lives in the repository and is reviewed before it is loaded, not pulled from an automatic feed. A company cannot appear or vanish because a data provider changed its mind overnight.

What is not in it

• Foreign secondary listings. Toyota trades in London as TYT.L, Panasonic as 0QYR.L, Nintendo as 0R1E.L. Of the London listings above the floor, fewer than half are actually UK companies; the rest are lines like these. They are excluded because the same company appearing twice under two tickers would give it two scores and split its dividend record, and because a depositary receipt's dividend behaviour reflects the receipt's mechanics as much as the company's policy.
• REITs and BDCs. These are the largest omission and the one most people notice. They are scored on a different model on the income vehicles page, because dividend cover for a REIT means cover from AFFO rather than from earnings, and running both through one score would produce two numbers that look comparable and are not. The screener counts how many REITs and BDCs match whatever filter you have set, and links across, rather than pretending they do not exist.
• Funds, ETFs and investment trusts. Those have their own lane, covered by the ETF methodology.
• Companies below the cap floor. Small-cap dividend records are thinner, the data behind them is patchier, and a cut risk score built on patchy data is worse than no score.

If a company you hold is missing, it is usually a secondary listing or a REIT.

Listings that have stopped trading

A listing whose most recent price is more than ten days old is not shown anywhere on the screener.

This matters more than it sounds. When a company is taken over, delisted, or simply drops out of the data provider's coverage, the price stops updating but the dividend record does not vanish. The yield calculated from the two carries on looking perfectly ordinary: one London listing in the universe still reports a yield of 5.6% from a share price last seen in March.

Nothing else on the row gives it away. The refresh timestamp says the row was checked last night, because it was. The data quality flag describes the inputs to the score, not whether the company still trades. So the last traded date is recorded separately and checked on its own.

Ten days is the threshold. It is long enough to cover a Christmas or Easter closure without dropping a company that is trading normally, and short enough that a real suspension is caught in a fortnight rather than a quarter. 52 of the 1,258 companies in the universe are currently excluded by this rule. Some were acquired, some delisted, and for a few the data provider simply stopped covering them. From where we sit those cases look identical, and in all of them we cannot tell you what the company yields today, which is the only thing this page is for.

The exclusion is not permanent. If prices start arriving again the company reappears on its own.

The figures in the table

Yield is the trailing twelve months of declared dividends per share divided by the current price. It is measured per company as the nightly refresh reaches it, not all at once, so two rows can carry yields measured a day apart. The column header states the most recent measurement date on the page.

That matters for sorting. A day of price movement is usually under 2%, so a company shown at 5.0% might be 4.9% today. It does not change who is roughly at the top, but it means the ordering is partly a function of measurement time, and a yield sort is better read as a grouping than a ranking.

Payout is the dividend as a share of earnings. Above 100% means the company is paying out more than it earns, which it can do for a while from reserves or borrowing, and not forever.

FCF cover is free cash flow divided by dividends paid. Below 1.0x means the cash the business generated did not cover what it paid out. Both of these sit on the row rather than behind a click, because a high yield the company cannot afford is the single most useful thing to know and a 0 to 100 score cannot tell you which input drove it.

Cut risk runs from 0 to 100 and lower is better. It reads payout cover, debt, interest cover and dividend record together, and it is backtested against dividends that were actually cut; the dividend safety proof page shows how it performed. A score of 30 or below is what this site calls low risk, and that cutoff is a choice rather than a verdict on anything above it.

Growth measures how fast a dividend has actually risen and how consistently, not how fast it might. A lower yield growing steadily can compound past a higher one that stands still.

Size is market capitalisation, shown in the listing's own currency and ranked in dollars. Those are deliberately two different things: showing a UK company's size in dollars would be a wrong number on the page, and ranking a pound figure against a dollar figure would put a bigger company second.

A dash means the figure has not been measured yet, never that it is zero. Around 1,100 of the companies in the universe are still waiting for their first nightly pass, and they fill in as it works through.

What the scores do not do

They do not predict the share price. They do not know about a takeover, a rights issue or a profit warning announced this morning. They read filed accounts and dividend history, which are backward-looking by construction, and they are wrong sometimes in both directions.

The dividend safety proof page is the honest version of how often: it shows what the cut risk score scored before each dividend cut in the backtest, including the ones it missed.