Methodology › Dividend Value Score
Dividend Value Score: How It Is Calculated
The Dividend Value Score is a number from 0 to 100 that summarizes how a security's current valuation compares with its own history and with similar securities: where today's dividend yield sits in its own past range, how its price multiple compares with its own five-year median and with its peers, and how much cash flow its market value buys. This page sets out every input, table and weight.
It is not a judgement that a price is too high or too low. A high score means today's yield is high and today's multiples are low compared with the security's own past and its peers. That can happen because the business has become more profitable, or because its share price has fallen on bad news. Read it alongside the Dividend Safety Score, which looks at the support behind the dividend.
What the score answers
The score answers one question: how does the current valuation compare with the security's own history and with peers? Its largest part compares today's dividend yield with the yields the same security has had over up to ten years. The rest compares a price multiple with its own five-year median and with the median of similar securities, and, for most profiles, measures the cash-flow yield of the company's market value.
What the score is not:
- Not a probability. A score of 80 does not mean an 80% chance of anything.
- Not a forecast. It summarizes reported figures and the dividend record as they stand; it does not project them.
- Not advice. It is the same for every visitor and takes no account of anyone's circumstances, holdings or objectives. It is not a recommendation to buy, sell or hold any security, and it is not a credit rating.
- Not a complete picture. It uses only the measures listed on this page. Business prospects, management, competition, pending transactions, legal matters and anything else not in our data are not part of it.
Profiles
Businesses fund their dividends in different ways, so one set of ratios does not fit all of them. A bank's debt is its raw material, a real estate investment trust's earnings are reduced by property depreciation that does not consume cash, and a business development company holds loans valued at market prices. Each security is therefore assigned a profile, and each profile uses the measures that fit its kind of business. The profile comes from our classification of every security by type and kind of business, which is based on its listing, its filings with the SEC and our data providers' records (see Security classification).
For the Value Score the profile decides which price multiple is compared. Reported earnings describe some businesses better than others, so each profile is compared on the multiple that suits it:
| Profile | Multiple compared with own history and peers | Cash-flow yield | Other |
|---|---|---|---|
| Ordinary companies | Price ÷ earnings (P/E) | Free-cash-flow yield | not used |
| Utilities | Price ÷ earnings (P/E) | Free-cash-flow yield | not used |
| Insurers | Price ÷ book value (P/B) | not used | not used |
| Banks | Price ÷ book value (P/B) | not used | not used |
| REITs, MLPs | Price ÷ operating cash flow per share (P/OCF) | Operating-cash-flow yield | not used |
| BDCs | not used | not used | Price ÷ net asset value (NAV) per share |
| Funds, preferred shares and other securities | Not Rated (see below) | ||
Banks and insurers are compared on book value because their assets are mostly financial and their earnings swing with credit losses, claims and investment gains; for the same reason their cash flow is not used. A BDC is compared only on its yield history and on its price against its net asset value, so that net asset value counts once rather than twice. REITs and MLPs are compared on operating cash flow because depreciation makes their earnings understate the cash they generate.
Inputs, sources and periods
Prices come from Financial Modeling Prep's (FMP) daily price history, adjusted for stock splits. For the historical comparisons we keep the closing price at the end of each month for up to ten years. Dividends come from FMP's dividend history, processed by our dividend engine: regular dividends only, restated for stock splits so that they match the split-adjusted prices. Financial figures come from companies' financial statements as supplied by FMP.
| Input | How it is calculated | Period |
|---|---|---|
| Yield against own history (YH) | At each month-end, the trailing yield is the regular dividends with an ex-dividend date in the previous 12 months divided by that month's closing price. Today's trailing yield, on the latest close, is then ranked against that monthly series: a percentile of 70 means it is higher than in 70% of the months on record (months with an equal yield count half). Left out of the series are the first months of a run of payments (until a full year has been paid, including after a dividend resumes following a long gap), months with no regular payment in the previous 12 months (for example during a suspension), yields treated as data errors, and months before a break in the data provider's share units around a split or spin-off. With 24 to 35 months of history, today's trailing yield is instead divided by the average of the monthly yields over the last five years | Up to 10 years of month-ends (at least 24 months) |
| Multiple against own history (PEH) | Today's multiple divided by the median of the same multiple at each of the last five fiscal year-ends. Today's P/E is the latest closing price divided by diluted earnings per share over the trailing twelve months (or the latest fiscal year where no trailing figure is available); P/OCF uses operating cash flow per diluted share over the same period; P/B uses book value per share (common equity ÷ diluted shares) from the newest balance sheet. For each of the last five fiscal year-ends, the same multiple is calculated with the month-end close nearest that date and that year's figures; years with zero or negative earnings, cash flow or equity are skipped, and at least three usable years are needed for a median. Per-share figures are restated for stock splits | Today and the last 5 fiscal year-ends |
| Free-cash-flow yield (FY) | Free cash flow (operating cash flow − capital expenditure) ÷ market capitalization × 100. REITs and MLPs: operating cash flow ÷ market capitalization. Not used for banks, insurers or BDCs. Not available when the company reports in a different currency from the one its shares trade in | Cash flow: trailing twelve months. Market capitalization: latest quote |
| Price ÷ NAV (PNAV) | Latest close ÷ net asset value per share (common equity on the newest balance sheet ÷ shares outstanding) | Latest balance sheet |
| Multiple against peers (PEER) | Today's multiple divided by the median of the same multiple among its peers: the securities with the same profile and the same sector that are eligible for scores (current payers, not held back by a data check) and have a positive multiple, the security itself included. If that group has fewer than 5 members, or the security has no sector on record, the peer comparison is missing. Not used for BDCs | Today |
The multiples are not calculated for ADRs, whose financial statements are per ordinary share while the price is per depositary share, or when a company reports in a different currency from the one its shares trade in. For those securities the Value Score rests on its other components, and its completeness shows the gap.
The yield in this score is the trailing yield (dividends paid over the last 12 months), because it can be measured the same way at every past month-end. It can differ from the forward yield shown at the top of each stock page, which annualizes the latest declared dividend.
Scoring tables
Each measure is turned into points from 0 to 100 using the fixed table shown for it. Between two listed values the points change in a straight line, and beyond the first or last listed value they stay at that row's points. For example, on a table that gives 100 points at 40% and 80 points at 60%, a value of 50% receives 90 points. The tables are the same for every security in a profile; they do not depend on how other securities score, except where a measure is explicitly compared with peers.
Yield against own history
With 36 months of history or more, the points equal the percentile itself. With 24 to 35 months, the ratio table is used instead.
| Percentile rank of today's trailing yield within its own monthly history (0 = lowest, 100 = highest) | Points |
|---|---|
| 0 | 0 |
| 25 | 25 |
| 50 | 50 |
| 75 | 75 |
| 100 | 100 |
| Today's trailing yield ÷ average trailing yield over the available history | Points |
|---|---|
| 0.70 or less | 10 |
| 1.00 | 50 |
| 1.30 or more | 90 |
Multiples
| Current multiple ÷ its 5-year median | Points |
|---|---|
| 0.60 or less | 100 |
| 0.80 | 80 |
| 1.00 | 50 |
| 1.20 | 25 |
| 1.50 or more | 0 |
| Current multiple ÷ median of the same multiple among peers | Points |
|---|---|
| 0.60 or less | 100 |
| 0.80 | 75 |
| 1.00 | 50 |
| 1.25 | 25 |
| 1.60 or more | 0 |
| Share price ÷ NAV per share | Points |
|---|---|
| 0.80 or less | 100 |
| 0.90 | 80 |
| 1.00 | 60 |
| 1.10 | 35 |
| 1.25 or more | 0 |
Cash-flow yield
| Free cash flow ÷ market capitalization | Points |
|---|---|
| 0% or less | 0 |
| 2% | 25 |
| 4% | 55 |
| 6% | 80 |
| 8% or more | 100 |
| Operating cash flow ÷ market capitalization | Points |
|---|---|
| 0% or less | 0 |
| 4% | 20 |
| 7% | 50 |
| 10% | 80 |
| 13% or more | 100 |
Weights by profile
Each profile's weights add up to 100. A dash means the measure is not used for that profile.
| Component | Ordinary companies, utilities | REITs, MLPs | Banks, insurers | BDCs |
|---|---|---|---|---|
| Yield against own history (YH) | 40 | 40 | 45 | 40 |
| Multiple against own 5-year median (PEH) | 25 | 25 | 30 | not used |
| Free-cash-flow yield (REITs, MLPs: operating-cash-flow yield) (FY) | 20 | 20 | not used | not used |
| Price ÷ net asset value (PNAV) | not used | not used | not used | 60 |
| Multiple against peers (PEER) | 15 | 15 | 25 | not used |
| Total | 100 | 100 | 100 | 100 |
Missing data and completeness
Every Value Score is published with a completeness figure: the share of the score's total weight whose inputs were available for that security. When an input is missing, it is left out and the remaining weights are scaled up so that they again add to 100; a missing input is never treated as zero and never filled with an estimate. Completeness tells you how much of the formula actually had data behind it.
A score is published only when completeness is at least 60% and every required input is present. Otherwise the security shows Not Rated, with the reason.
Out-of-date statements are treated as missing. If the latest usable financial statements for a company cover a period that ended more than 15 months ago, every input that comes from them (payout ratios, leverage, interest coverage, growth in revenue and earnings, cash-flow yield) is treated as missing for scoring, so the score's completeness falls and it may be Not Rated. The stock page can still show those older figures, labelled with their fiscal year.
| Share of the score's weight with data | Label |
|---|---|
| 90% or more | Complete data |
| 75% to under 90% | Most data available |
| 60% to under 75% | Partial data |
| Under 60% | Not Rated |
Required input: the yield against its own history (YH). A security with less than 24 months of price and dividend history, for example a recent listing or spin-off, has no history to compare with and is Not Rated.
When a security is Not Rated
A security shows Not Rated, with the reason stored and shown, when any of these applies:
- It is a fund. ETFs, closed-end funds and fund or trust units: fund scores are not yet available. A fund's distributions depend on its holdings, its net asset value and, sometimes, return of capital, which these formulas are not built to assess.
- It is not a common share. Preferred shares, exchange-traded notes and baby bonds, units, warrants and rights: the scores apply to common shares. Payments on these securities are set by their terms (coupon, call date, ranking), which these measures do not capture.
- Its classification is incomplete. A security whose type, or whose issuer's kind of business, has not yet been classified cannot be assigned a profile, so it is not scored until it is.
- It is no longer covered. A security that has left our coverage keeps no current scores.
- It is not a current payer. It has paid no regular dividend in the last 12 months and has none declared and still to come; its only recent payments were one-off specials; its regular dividend has lapsed; its computed yield is treated as a likely data error; or a corporate action such as a spin-off makes its dividend record not comparable with today's shares. These are the same rules that decide whether a security shows a yield (see Yield calculation).
- An automated data check is holding it back. A check that marks one of its figures as likely wrong (shown as "Withheld" on our data quality page) withholds its scores until the check clears.
- A required input is missing. Fewer than 24 months of price and dividend history, so today's yield cannot be compared with its own past.
- Completeness is under 60%.
Score bands
The score is always shown as a number, with a short descriptive label. The labels summarize the score as a whole, which combines the comparison with the security's own history, the comparison with its peers (not for BDCs) and, for most profiles, its cash-flow yield. A higher score means a lower valuation on these measures. The labels are not a judgement that a price is too high or too low.
| Score | Label shown with the score |
|---|---|
| 80 to 100 | Much lower valuation than its own history and peers |
| 60 to 79 | Lower valuation than its own history and peers |
| 40 to 59 | In line with its own history and peers |
| 20 to 39 | Higher valuation than its own history and peers |
| 0 to 19 | Much higher valuation than its own history and peers |
Worked example
Illustrative only
The company below is hypothetical. Its figures are made-up round numbers chosen to show the arithmetic; they do not describe any real company.
The same hypothetical ordinary company used on the Safety Score page has a trailing yield today that is higher than in 70% of the month-ends of the last ten years. Its P/E is 18 against a five-year median of 20 (a ratio of 0.90, halfway between the table's 0.80 and 1.00, so 65 points), its free-cash-flow yield is 5%, and the median P/E of its peers (same profile and sector) is 16, so its ratio to peers is 1.125.
| Component | Hypothetical input | Points | Weight | Points × weight ÷ 100 |
|---|---|---|---|---|
| Yield against own history (YH) | 70th percentile of its 10-year range | 70.0 | 40 | 28.00 |
| P/E against own 5-year median (PEH) | P/E 18 ÷ median 20 = 0.90 | 65.0 | 25 | 16.25 |
| Free-cash-flow yield (FY) | 5% | 67.5 | 20 | 13.50 |
| P/E against peers (PEER) | P/E 18 ÷ peer median 16 = 1.125 | 37.5 | 15 | 5.62 |
| Total | 100 | 63.38 |
The weighted points add up to 63.38, so the Value Score is 63.4, "Lower valuation than its own history and peers", completeness 100% ("Complete data"). The comparison with its own history points one way and the comparison with peers the other; the score shows both, weighted as in the table above.
Updates and versioning
The Value Score is recalculated every day in our morning data refresh, before the U.S. market opens, from the latest stored inputs. Each input is only as fresh as the data it comes from: prices, the dividend calendar and newly declared dividends refresh daily; financial statements refresh weekly, so a newly published quarterly report usually reaches the score within about a week of our data provider publishing it. The full schedule is on the methodology page, and the last refresh of each data set is shown on our data quality page. A daily snapshot of every score is kept, so changes over time can be traced.
The historical series behind the comparisons uses month-end closing prices, so each month adds one point to it.
The formula version in use is 2026-10-19.4, introduced on October 19, 2026. The version is stored with every Value Score, and any change to a curve, weight, rule or input bumps it. Each change is listed here with its date, so a score can always be matched to the formula that produced it. Scores calculated under different versions are not directly comparable.
| Version | Date | Change |
|---|---|---|
| 2026-10-19.4 | October 19, 2026 | Financial statements whose period ended more than 15 months ago are no longer scored: the score treats them as missing, so its completeness falls and it may be Not Rated. Dividend-record fixes: a stock split dated on or the day before an ex-date no longer reads as a dividend cut when the payment was already quoted in post-split shares, and a switch from quarterly to monthly payments no longer reads as a cut. |
| 2026-10-19.3 | October 19, 2026 | REITs and MLPs: interest coverage measured on EBITDA (before depreciation). Banks: revenue growth no longer used in the Growth Score, because the revenue reported for banks is mostly gross interest income. S.A.F.E. Solid now requires a Safety Score of 80 or higher (was 70), because most U.S.-listed common stocks scored above 70. |
| 2026-10-19.2 | October 19, 2026 | Methodology review: a dividend cut counts against the dividend record only when the regular rate falls by 20% or more; earnings stability measured as a share of the years available, so a short history is not capped; earnings per share that is negative at the end of three years scores 0 instead of being left out; REITs and MLPs no longer fall back to revenue for earnings growth; the operating-cash-flow yield table starts from 0; insurers valued on book value like banks; BDCs valued on yield history and price to net asset value only; negative equity scores 0 on debt to equity; new Value Score labels. |
| 2026-10-19.1 | October 19, 2026 | First version of the Dividend Safety, Dividend Growth and Dividend Value Scores. DividendRank becomes the published composite of the three scores, replacing the earlier formula (yield, payout ratio, return on equity, price-to-book and length of payment history). The S.A.F.E. Solid check uses the Dividend Safety Score instead of DividendRank. |
Validation
Validation: a point-in-time test of the Safety Score against later dividend cuts is in progress; results will be published here when complete.
The test will compute the score as it would have stood at past dates, using only data available at each date, and compare it with the dividend cuts and suspensions that followed. Until results are published, no claim is made about how well any score has anticipated dividend changes.
The Value Score is not part of that test.
Known limitations
- A security's own past is not always a fair reference. After a merger, a spin-off, a change of business or a change in interest rates, its earlier yields and multiples may describe a different situation.
- A high yield against history has more than one cause. It may follow a rising dividend or a falling share price, and a falling price can reflect real problems in the business. The score does not tell these apart; the Safety Score and the stock page's dividend history help.
- Peer groups can be small or mixed. Peers share a profile and a sector; a group needs at least 5 members, but a median of 5 still rests on few companies, and a sector can contain quite different businesses.
- ADRs and foreign-currency reporters have fewer inputs. Their multiples are not calculated, so their Value Score rests mostly on the yield comparison.
- Multiples are distorted by one-off items. One-time gains and charges in earnings, and years with losses (which are skipped), affect P/E comparisons.
- Trailing, not forward. The yield compared is the trailing 12-month yield, which lags a newly declared increase or cut until the new payments are made.
- History is limited to 10 years and to what our data provider supplies; foreign listings and recent listings often have less.
- The tables and weights are fixed judgments. They are Dividendly's own choices, set in advance, and have not been fitted to later returns or dividend outcomes.
Related pages
- Methodology overview: yields, specials, growth and streaks, payout ratios, classification and refresh schedule
- Dividend Safety Score
- Dividend Growth Score
- DividendRank and S.A.F.E. checks
- Data sources and data quality and corrections
- Editorial policy and disclaimers
- Found a figure that looks wrong? Report an error.
Research only. Dividendly's scores describe securities, not people: they are produced by the published formulas on these pages and are the same for every visitor. They are not investment advice and not a recommendation to buy, sell or hold any security. Data may be delayed, incomplete or wrong; see our disclaimers.