Why Two Financial Platforms Show Different Numbers for the Same Company

You pull up the same company on two financial platforms and the numbers do not match. Revenue is $51.2 billion on one platform and $52.0 billion on another. EPS is $2.14 in one place and $2.08 somewhere else. Market capitalization, valuation multiples, and profitability ratios can differ too.

The immediate assumption is that one platform has a data error. That is not necessarily what happened. Two sources can rely on the same public disclosures yet display different values because they use different reporting periods, source versions, share-count conventions, currencies, timestamps, standardized fields, or calculation methods. The same issue can arise when you compare a current platform value with an older spreadsheet or model.

The useful question is not simply which number is correct. It is whether the two values are measuring the same thing. This guide helps you determine whether two values are actually comparable before treating either source as wrong. It is not a provider scorecard, a filing-mapping audit, or a complete data-governance framework.

Compare the context around a metric before treating two reported values as a data conflict.

Key takeaways

• A difference between platforms does not automatically indicate bad data. First compare the security, metric definition, period, currency, source version, and underlying inputs.
• Revenue and earnings require period and reporting-basis checks. Market capitalization and enterprise value also depend on market-data observations and capital-structure inputs.
• Ratios should be reconciled from their numerator, denominator, timing, and methodology rather than compared only at the final percentage or multiple.
• A current endpoint response can provide the current available record, but it does not recreate what a platform displayed at an earlier point in time.
• The strongest model preserves the context around an important value, not only the final number.

Start by establishing what each value represents

Before comparing the number itself, compare its context.

Check	Why it matters
Issuer and security	A company name alone may not identify the same share class, listing, or reporting entity.
Metric definition	“Revenue,” “EPS,” or “market cap” can reflect different presentations or calculation conventions.
Reporting period	Quarterly, annual, and trailing-twelve-month values are not interchangeable.
Period-end date	A shared fiscal-quarter label can still represent different calendar dates.
Currency and units	A reported EUR value and a USD-converted value can differ without an accounting change.
Source version	An original filing, amendment, or later standardized record may not match.
Observation date	Price-linked metrics depend on the market-data date and, where supplied, observation time.
Calculation inputs	Derived ratios inherit differences in the inputs and methodology beneath them.

A mismatch becomes meaningful only after those fields have been aligned.

The reporting timeline often explains the difference

Financial data has more than one relevant date. A company reports results for a defined period, files or publishes those results, and may later amend or revise them. A data provider then processes the record and makes it available to users.

Those steps do not necessarily occur at the same time across platforms.

Consider an illustrative company that originally reports annual revenue of $100.0 billion. An amended filing later revises the figure to $99.4 billion.

Field	Platform A	Platform B
Issuer	Same company	Same company
Fiscal year	FY2025	FY2025
Period end	January 31, 2025	January 31, 2025
Reported currency	USD	USD
Source version	Original filing	Amended filing
Displayed revenue	$100.0B	$99.4B
Data observed	February 3, 2025	March 12, 2025

The platforms are not necessarily disagreeing about the same record. They may be showing different versions of the same historical period.

For price-linked metrics, the observation date matters as well. A company with 2 billion shares outstanding has an implied market capitalization of about $100 billion at $50 per share. At $49.60 per share, the implied market capitalization is about $99.2 billion even though its financial statements have not changed.

FMP's Company Profile Data API provides current company, security, price, market-capitalization, and identifier context. Its Income Statement, Balance Sheet Statement, and Cash Flow Statement APIs provide dated financial-statement records with fields such as period end, fiscal year, fiscal period, filing date, and reported currency.

A financial value should therefore be treated as a value with context, not as a permanent fact detached from time.

Fiscal-period alignment changes what the number means

A label such as “Q1,” “FY2026,” or “TTM” is not sufficient for a direct comparison.

Companies use different fiscal calendars. A retailer with a January year-end and a company with a December year-end can both report “Q1 2026,” even though the underlying operating periods cover different months.

Trailing-twelve-month calculations create another common mismatch. Suppose quarterly revenue is $8 billion, $9 billion, $10 billion, and $11 billion. The reported annual value is $38 billion. After the next quarter reports $12 billion, a trailing-twelve-month calculation replaces the oldest $8 billion quarter with the new $12 billion quarter, producing $42 billion.

Both values can be valid. One is an annual reported result; the other is a rolling four-quarter measure.

FMP's Key Metrics API supports period-based metrics, while the Key Metrics TTM API provides trailing-twelve-month measures. Do not compare one with the other unless the analytical question calls for that difference.

Metric	What must match before comparison
Quarterly revenue	Fiscal period and period-end date
Annual revenue	Fiscal year and reporting basis
TTM revenue	Exact rolling four-quarter window
Quarterly EPS	Fiscal period and EPS basis
Annual EPS	Fiscal year and share-count treatment
Margin or return metric	Period plus numerator and denominator definition

The period end, announcement date, filing date, and data-observation date answer different questions. Treating them as interchangeable can attach an otherwise valid number to the wrong analytical period. Keeping those records separate is especially important when reported results, estimates, and market data appear in the same model.

Standardized data can differ from a company's presentation

Companies do not publish financial statements in one universal database-ready structure. They use their own labels, classifications, and levels of detail. Financial-data providers then organize those disclosures into standardized fields so analysts can compare companies and periods more consistently.

That standardization is useful, but it creates a distinction between what a company reported and how a platform presents it.

One source may group items together while another keeps them separate. One may display a standardized revenue field while another preserves an issuer-specific label. The difference is not automatically an error.

If the question is whether a normalized field maps cleanly to the company's original filing, compare the standardized and as-reported statements before treating the difference as a mapping issue. That is a deeper source-mapping task than this article's cross-platform comparability check.

FMP provides standardized statement data through its financial-statement APIs and filing-oriented data through the As Reported Financial Statements API. Use the distinction deliberately.

The same principle applies to ratios. If net income is $500 million and ending equity is $2 billion, return on equity is 25%. If the calculation uses average equity of $1.8 billion, return on equity is about 27.8%.

Neither result is useful without knowing the denominator methodology.

The Financial Ratios API and Key Metrics API provide structured derived measures, but a ratio comparison still requires the analyst to confirm the period, inputs, and definition.

EPS can differ even when net income is the same

EPS appears simple because it is a per-share measure. The earnings figure and share-count treatment can still produce different results.

Suppose a company reports $1 billion of net income.

• At 500 million diluted weighted-average shares, diluted EPS is $2.00.
• At 510 million diluted weighted-average shares, diluted EPS is about $1.96.

Net income has not changed. The denominator has.

A cross-platform EPS comparison should establish:

• Whether the figure is basic or diluted EPS
• Whether the earnings figure is reported GAAP or a company-defined adjusted measure
• The weighted-average share count used for the period
• Whether discontinued operations or other earnings components are included
• Whether historical per-share values were adjusted for a stock split
• The exact fiscal period

The Income Statement API exposes reported EPS, diluted EPS, and weighted-average basic and diluted share fields. Those fields help you inspect the reported components instead of debating the final P/E ratio first.

Market capitalization and enterprise value depend on different inputs

Market capitalization is a market value, not a financial-statement line item.

In basic terms:

Market capitalization = Share price × Shares outstanding

The relevant share count is shares outstanding as of the market-data observation. It is not the weighted-average basic or diluted share count used to calculate EPS.

If a company trades at $50 and has 2 billion shares outstanding, its market capitalization is about $100 billion. If another source uses 1.96 billion shares, the same price produces a market capitalization of about $98 billion.

Historical market-cap comparisons should also match the observation date. Use an exact intraday time only when the underlying source provides one; otherwise compare the available date-based observations. FMP's Company Market Cap API supports historical market-cap analysis by date.

Enterprise value adds more inputs. A simplified framework is:

Enterprise value = market capitalization + debt − cash

Providers can still differ in their treatment of debt, cash, leases, preferred equity, noncontrolling interests, or EBITDA. Compare the included components rather than assuming two displayed enterprise values use an identical convention.

The Enterprise Values API provides enterprise-value-related fields alongside market capitalization, debt, and cash-related information.

Currency can create an apparent mismatch without an accounting change

A company may report revenue in euros while one platform displays the reported amount and another converts it into U.S. dollars.

Suppose reported revenue is €10 billion:

• At 1.10 USD per EUR, the converted value is about $11.0 billion.
• At 1.08 USD per EUR, the converted value is about $10.8 billion.

The underlying reported revenue has not changed. The conversion assumption has.

The timing convention matters too. Income-statement values represent activity over a period, while balance-sheet values represent a position at a point in time. A conversion method appropriate for revenue may not be appropriate for year-end cash or debt.

When a model uses converted values, preserve both the reported value and the conversion assumption.

Field	Example
Reported currency	EUR
Reported revenue	€10.0B
FX assumption	1.10 USD per EUR
Converted revenue	$11.0B

That structure keeps the original company-reported amount visible and makes the translation method auditable.

Restatements can change the current historical record

A spreadsheet created months ago can legitimately differ from a platform's current historical value.

A company may amend a filing, reclassify an item, include an audit adjustment, or formally restate a prior period. When a provider incorporates the later record, the current displayed value can differ from what was available when the original model was built.

For current analysis, the latest available historical record may be the appropriate baseline. For point-in-time research, that same value may be unsuitable if it was not available on the date the historical decision would have been made.

A current endpoint response does not by itself reconstruct what a platform displayed on an earlier date. Preserve the supporting filing or source version and your own observed_at snapshot when point-in-time accuracy matters. A revised input can also flow through ratios, margins, and valuation measures.

Reconcile two platforms without guessing

Start with the one metric that differs. Do not compare an entire company page and assume every mismatch has the same cause.

For revenue, begin with reporting period and source version. For EPS, add the earnings basis and weighted-average share count. For market capitalization, focus on shares outstanding and the market-data observation. For a ratio, trace the numerator and denominator back to the inputs and calculation method.

Check	What to compare across platforms
Issuer and security	Confirm that both values refer to the same reporting entity and security, including the listing and share class.
Metric definition	Check what each field measures. Matching labels do not necessarily mean matching definitions.
Fiscal period and period end	Confirm the reporting window and its end date. Distinguish quarterly, annual, and trailing-twelve-month values.
Currency and units	Check the currency, any conversion assumptions, and whether values are expressed in units, thousands, millions, or billions.
Source presentation	Determine whether each platform uses standardized data or the company's as-reported presentation.
Earnings basis, where relevant	Check whether the figure uses reported GAAP earnings or a company-defined adjusted measure.
Corporate-action treatment	For historical per-share values, check whether both sources apply the same stock-split adjustments.
Filing or record version	Identify whether each value comes from the original filing, an amendment, or a subsequently revised record.
Market-data observation, where relevant	Match the observation date for price-linked inputs. Compare intraday times only when the sources provide them.
Inputs and methodology	Compare the underlying calculation components, including the numerator, denominator, share-count treatment, and timing.

The reconciliation should end with one of three outcomes:

1. Same metric, different observation or calculation inputs. The difference is explained and can be documented.
2. Different metric or reporting basis. The values are not directly comparable until the intended basis is selected.
3. Unresolved. The available evidence does not establish the reason for the difference, so the underlying filing or source record needs review.

Illustrative reconciliation: original versus amended revenue

Suppose two platforms show different annual revenue for the same company.

Check	Platform A	Platform B	Result
Issuer and security	Same issuer	Same issuer	Match
Fiscal year and period end	FY2025; January 31	FY2025; January 31	Match
Currency	USD	USD	Match
Metric definition	Reported annual revenue	Reported annual revenue	Match
Source version	Original filing	Amended filing	Different
Displayed value	$100.0B	$99.4B	Explained

The outcome is not that one platform is automatically wrong. It is that both values refer to the same metric and period, but not the same version of the source record.

When a recurring model has an unresolved discrepancy, turn it into a pre-model validation check rather than resolving it ad hoc each time.

Preserve a usable baseline in your model

A defensible model keeps the context beside the value:

• Metric definition
• Issuer and security identifier
• Fiscal period and period-end date
• Reported currency and units
• Source presentation and accounting basis
• Filing or source version
• Market-data observation date for price-linked measures
• Calculation inputs and methodology
• Retrieval or observation time

Structured data makes those checks easier to repeat. The analytical choice remains yours: select the value that fits the purpose of the model, document why it was selected, and do not let a familiar label conceal a different underlying measure.

Frequently asked questions

Why do two financial platforms show different revenue for the same company?

The platforms may use different fiscal periods, annual versus TTM values, source versions, classifications, currencies, or historical revisions. Confirm that both values represent the same company, period, currency, metric definition, and filing version before treating the difference as an error.

Why is EPS different across financial websites?

EPS can differ because platforms use different earnings definitions, basic versus diluted shares, weighted-average share counts, fiscal periods, adjusted earnings, or split-adjustment conventions. Compare the earnings numerator and share-count denominator before comparing the final EPS value.

Why does market capitalization differ between platforms?

Market capitalization depends on share price and shares outstanding at the relevant market-data observation. A difference in either input, including a different observation date, can produce a different market-cap figure.

Why can two platforms show different P/E ratios?

P/E depends on both price and EPS. Even with the same price, the ratio can differ if one source uses basic, diluted, adjusted, annual, or TTM earnings while another uses a different basis.

Which number should go into a financial model?

Use the value that matches the purpose of the model, then preserve the context that makes the choice reproducible. Record the metric definition, fiscal period, source presentation, currency, filing or record version, market-data observation where relevant, and calculation methodology.