DividendMapper

Dividend safety, tested

Know which dividends to trust.

The average dividend payer cuts about 1 in 5 times. The ones we rate high risk cut about 1 in 2. This page shows you the working.

4,184 real dividend cuts1,767 US and UK companies2015 to 2024No card needed

The risk it catches

1 in 2

Holdings we rate high risk cut their dividend within a year. The average payer cuts about 1 in 5 times.

High risk (90+)1 in 2
Average payer1 in 5
Low risk (under 20)1 in 7

Cut within 12 months, measured on years the engine was never tuned on.

The income it protects

6 in 7

Holdings we rate low risk kept paying, through 2020 and 2024.

Proof, not promises

The worse the score, the more dividends got cut.

Every band, flattering and not. Checked against 4,184 real dividend cuts across 1,767 US and UK companies from 2015 to 2024, including a stretch the engine was never shown.

Risk scoreCut within a yearHoldings tested

Under 40

1 in 713%8,547

40 to 59

1 in 616%1,885

60 to 79

1 in 334%1,097

80 to 89

1 in 338%380

90 and over

1 in 253%646

Every payer, for comparison

1 in 518%

These are dividend-cut rates, not share-price forecasts. Measured on the years the engine was never taught from, so it cannot have marked its own homework.

We changed how a cut is counted in August 2026, and it moved these numbers in our favour, so here is why. The old method compared each dividend payment against the ones before it. That reads correctly for a company paying four equal amounts a year, and badly for most UK companies, which pay one small dividend and one large one. A company that never cut could look like it cut every year, and a company that stopped paying altogether could look untouched. We now compare a full year of dividends against the year before it. The score itself did not change. The question we ask of it did.

Why you can trust it

Marked on years it had never seen.

Anyone can look clever explaining the past. We split ten years of dividend history three ways, and a change only ships if it still works on the years it was never taught.

2015 to 2018

Where it learns

The engine studies what a dividend looks like before it gets cut.

2019 to 2023

Where it proves itself

Every change has to work here, on years it did not learn from.

2024

The final exam

Kept aside and never allowed to look at until the very end. If a change fails here, it does not ship.

What the score does not do

It flags the risk of a dividend being cut. It does not predict share prices, and it is not financial advice or a recommendation to buy or sell. It is a risk screen on the income you already own. See the full methodology and data.

It keeps testing itself

The score is never finished.

Week after week it tries small improvements on itself, marks each one against the years it was never taught, and throws away the ones that do not hold up. A person makes the final call before anything reaches your dashboard.

21

Ideas tried

0

Good enough to ship

So far none were good enough. That is the bar doing its job. We would rather ship nothing than ship a change that only looks good on the years it already knows.

On your dashboard

Every holding, scored.

Connect your broker or paste a CSV, and every holding gets a Risk score. The redder the pill, the more likely that income is to be cut. The shaky ones stand out without you going looking for them.

Holdings shown here are illustrations, not recommendations.

dividendmapper.com/app/portfolio
HoldingDividend safety

Utility Co

6.8% yield

Bank plc

4.1% yield

Staple Inc

3.2% yield

Questions, answered.

Is this financial advice?

No. The dividend-safety score is a risk screen on income you already own. It is not advice and not a recommendation to buy or sell. Always do your own research.

What data is the score built on?

Company fundamentals, cash flow and dividend history for US and UK large-cap dividend payers, going back to 2015. The proof numbers come from 4,184 real dividend cuts across 1,767 companies.

What does the score actually mean?

It ranks how likely a holding is to cut its dividend in the next year. Holdings scoring 90 and over cut about 1 in 2 times; holdings under 20, about 1 in 7.

How often does it update?

Scores refresh daily. The engine behind them keeps testing and improving itself, and only changes that beat a held-back 2024 test get promoted.

Is there a free option?

Yes. You get two free scores a day without an account, and the Pro trial runs 14 days with no card. It drops to the Free plan when it ends, nothing to cancel.

Which markets are covered?

US and UK large-cap dividend payers, with broker sync so you can score your real portfolio.

See which of your holdings are safe.

Score every holding you own and stop being surprised by a cut.

No card needed. Drops to the Free plan when it ends, nothing to cancel.