American Express is a credit card company from New York. Unlike Visa or Mastercard, it issues the cards itself and also runs the payment network. It earns merchant fees, annual fees and interest. Customers are mainly high-income individuals and businesses. As of Oct 11, 2026
American Express is a global payments company based in New York and was founded in 1850. The company issues credit and charge cards to consumers, small and mid-sized businesses and large corporations. Merchants pay American Express a fee for accepting these cards. Further revenue comes from interest on card loans and from the annual fees of many cards. The company runs its own global payment network; in about 110 countries and territories, partner banks issue American Express-branded cards or sign up merchants for the network. Its offering also includes deposits, non-card loans and security tools for merchants.
Automatically summarized from SEC annual report (10-K, filed Feb 6, 2026) (website of the US Securities and Exchange Commission) · reviewed Oct 15, 2026
Our own calculation from the exchange’s daily candles, as of Oct 26, 2026. “All-time” applies since the start of this series, not to the whole market. Changes in percent; the TradingView chart shows prices. “Today” = change of the last trading day versus the previous day. Not investment advice.
Still above the buy zone
After a rise there is a buy zone; the price is still above it. “Near the buy zone” (from Watcher) shows how close it is.
As of: Close of the 4-hour candle Oct 27, 2026, 18:30 · Buy zone in the trade plan · An observation, not investment advice
Rule (4-hour candles) Zone valid since Oct 23, 2026 · unchanged since then Zone historyThe zone history shows since when a buy zone has applied, when it was adjusted and when the price touched or left it. Recording began on Oct 12, 2026, only runs forward and fills nothing in – older zones are marked “known at start of recording” or “already existing when added”.Example: New buy zone on Oct 14, no adjustment since → “zone valid since Oct 14, 2026 · unchanged since”.More in the Academy: Lesson buy zone and Fibonacci
Watcher Full history: every change of the buy zone with its date.
Weekly candles (1W) Zone known when recording started (recording since Oct 12, 2026) · Setup confirmed on Jul 20, 2026 · unchanged since recording started Zone historyThe zone history shows since when a buy zone has applied, when it was adjusted and when the price touched or left it. Recording began on Oct 12, 2026, only runs forward and fills nothing in – older zones are marked “known at start of recording” or “already existing when added”.Example: New buy zone on Oct 14, no adjustment since → “zone valid since Oct 14, 2026 · unchanged since”.More in the Academy: Lesson buy zone and Fibonacci
Watcher Full history: every change of the buy zone with its date.
Timeframe 1 week · As of Oct 23, 2026 · display only, the rule works on 4-hour candles
Elliott count: Combination (W-X-Y), Corrective pattern (upward), Y complete (3 of 3 guidelines) · X = 80% of W; Y = 123% of W
Comparison: The rule treats the last rise as the possible start of an upward impulse (without checking its history and structure). The Elliott count reads it as the end of an upward correction (Y).
Alternative (not drawn): Zigzag (A-B-C), Corrective pattern (upward), C complete (1 of 2 guidelines)
Note: Structure ambiguous: leg W, Y can also be read as three- or five-part.
Note: Retracement at least 88.7% of A: flat possible (B in progress, type only with C).
Elliott count calculated automatically by fixed Elliott rules, not investment advice – display only, our buy zones do not use it. Provisional: The count can change with new candles.
Automated analysis (rule-based): the rule calculates waves and Fibonacci levels from the completed weekly candles; the chart is recalculated every day. Not investment advice. On this timeframe the chart only shows the rule setup (swing detection 2 × ATR) – the rule itself works on 4-hour candles. Axis and values: distance from the top of the buy zone in %, no prices. Fibonacci measured logarithmically (weekly candles).
On phones the drawing toolbar is hidden – drawing works better on a desktop.
Timeframe 1 day · As of Oct 26, 2026 · display only, the rule works on 4-hour candles
Elliott count: Flat (type known only after C), Corrective pattern (upward), C in progress (2 of 3 guidelines) · B = 161% of A; C = 85% of A
Note: Flat type (regular, expanded, running) can only be decided once C is complete.
Note: Structure ambiguous: leg B can also be read as three- or five-part.
Elliott count calculated automatically by fixed Elliott rules, not investment advice – display only, our buy zones do not use it. Provisional: The count can change with new candles.
No buy zone of the rule on this timeframe – the chart shows only the Elliott count from the completed daily candles (recalculated every day). Not a recommendation, not investment advice. Without a buy zone there is no reference for percentages – so no value axis and no prices; measurements are ratios of wave lengths.
Timeframe 4 hours · As of Oct 27, 2026
Elliott count: Combination (W-X-Y), Corrective pattern (downward), Y complete (3 of 3 guidelines) · X = 76% of W; Y = 123% of W
Alternative (not drawn): Zigzag (A-B-C), Corrective pattern (downward), C complete (2 of 2 guidelines)
Note: Structure ambiguous: leg W, Y can also be read as three- or five-part.
Elliott count calculated automatically by fixed Elliott rules, not investment advice – display only, our buy zones do not use it. Provisional: The count can change with new candles.
Automated analysis (rule-based): the rule calculates waves and Fibonacci levels from the completed 4-hour candles; the chart is recalculated every hour. Not investment advice. Axis and values: distance from the top of the buy zone in %, no prices.
On phones the drawing toolbar is hidden – drawing works better on a desktop.
Timeframe 1 hour · As of Oct 27, 2026 · display only, the rule works on 4-hour candles
Elliott count: Zigzag (A-B-C), Corrective pattern (upward), C complete (2 of 2 guidelines) · B = 56% of A; C = 99% of A
provisional count The inner sub-waves cannot be seen on this timeframe – the hard Elliott rules are met, the internal structure remains open. Provisional – check the count yourself on the chart.
Comparison: The rule treats the last rise as the possible start of an upward impulse (without checking its history and structure). The Elliott count reads it as the end of an upward correction (C).
Note: C is an ending diagonal.
Note: Structure open: leg A cannot be broken down into sub-waves on this timeframe – provisional, check the count yourself on the chart.
Elliott count calculated automatically by fixed Elliott rules, not investment advice – display only, our buy zones do not use it. Provisional: The count can change with new candles.
Automated analysis (rule-based): the rule calculates waves and Fibonacci levels from the completed hourly candles; the chart is recalculated every hour. Not investment advice. On this timeframe the chart only shows the rule setup (swing detection 4 × ATR) – the rule itself works on 4-hour candles. Axis and values: distance from the top of the buy zone in %, no prices.
On phones the drawing toolbar is hidden – drawing works better on a desktop.
Timeframe 1 week · As of Oct 23, 2026 · from the Elliott count in the analysis chart
Weighting Main scenario: ahead (3 of 3 guidelines, strong) · Alternative: behind (1 of 2, medium) Guidelines (strength of the count)Guidelines are typical features of a count, such as usual Fibonacci lengths or a change in the form of correction. Strong means: at least three quarters met, medium: at least half, otherwise weak. Even a strong count can flip.Example: 3 of 3 guidelines → strong; 2 of 4 → medium; 1 of 3 → weak.More in the Academy: Lesson wave degrees
Weighting by guidelines met – not a probability.
As long as the weekly close stays below the end of Y, this applies: complex countermove upward (Combination W-X-Y), Y complete – count points to falling prices.
Invalid if the weekly close is above the end of Y (+12.5% from the top of the zone)
If that breaks, the second reading no longer applies either: sharp countermove upward in three waves (Zigzag A-B-C), C complete – then there is no valid count for now; the next candles will produce a new one.
Invalid if the weekly close is above the end of C (+12.5% from the top of the zone)
Timeframe 1 day · As of Oct 26, 2026 · from the Elliott count in the analysis chart
Weighting Main scenario (2 of 3 guidelines, medium) · no alternative Guidelines (strength of the count)Guidelines are typical features of a count, such as usual Fibonacci lengths or a change in the form of correction. Strong means: at least three quarters met, medium: at least half, otherwise weak. Even a strong count can flip.Example: 3 of 3 guidelines → strong; 2 of 4 → medium; 1 of 3 → weak.More in the Academy: Lesson wave degrees
Weighting by guidelines met – not a probability.
As long as the daily close stays above the end of B, this applies: flat countermove upward in three waves (Flat, type known only after C), C in progress – count points to rising prices.
Invalid if the daily close is below the end of B
If that breaks, this count no longer applies. Currently no alternative that follows the rules.
Watcher Fibonacci target zones and levels for this count – as a help for your own review – from Watcher.
Timeframe 4 hours · As of Oct 27, 2026 · from the Elliott count in the analysis chart
Weighting Main scenario: ahead (3 of 3 guidelines, strong) · Alternative: behind (2 of 2, strong) Guidelines (strength of the count)Guidelines are typical features of a count, such as usual Fibonacci lengths or a change in the form of correction. Strong means: at least three quarters met, medium: at least half, otherwise weak. Even a strong count can flip.Example: 3 of 3 guidelines → strong; 2 of 4 → medium; 1 of 3 → weak.More in the Academy: Lesson wave degrees
Weighting by guidelines met – not a probability.
As long as the 4-hour close stays above the end of Y, this applies: complex countermove downward (Combination W-X-Y), Y complete – count points to rising prices.
Invalid if the 4-hour close is below the end of Y (−3.6% from the top of the zone)
If that breaks, the second reading no longer applies either: sharp countermove downward in three waves (Zigzag A-B-C), C complete – then there is no valid count for now; the next candles will produce a new one.
Invalid if the 4-hour close is below the end of C (−3.6% from the top of the zone)
Timeframe 1 hour · As of Oct 27, 2026 · from the Elliott count in the analysis chart
Weighting Main scenario (2 of 2 guidelines, strong) · no alternative Guidelines (strength of the count)Guidelines are typical features of a count, such as usual Fibonacci lengths or a change in the form of correction. Strong means: at least three quarters met, medium: at least half, otherwise weak. Even a strong count can flip.Example: 3 of 3 guidelines → strong; 2 of 4 → medium; 1 of 3 → weak.More in the Academy: Lesson wave degrees
Weighting by guidelines met – not a probability.
provisional count The inner sub-waves cannot be seen on this timeframe – the hard Elliott rules are met, the internal structure remains open. Provisional – check the count yourself on the chart. No visible sub-waves in: A.
As long as the hourly close stays below the end of C, this applies: sharp countermove upward in three waves (Zigzag A-B-C), C complete – count points to falling prices.
Invalid if the hourly close is above the end of C (+2.9% from the top of the zone)
If that breaks, this count no longer applies. Currently no alternative that follows the rules.
The price is still above the buy zone (0 to −1.6% from its top) (61.8–78.6% of the wave 1 rise). Rule trigger: the daily low reaches the zone, then a swing high forms and a daily close is above it. Whether the price reaches the zone is open – if it rises above the wave 1 high first, the setup is void. All percentages: distance from the top of the buy zone.
In the TradingView chart: Draw the buy zone in the chart with the Fibonacci tool (instructions next to the wave chart) or as two horizontal lines (tool “Horizontal Line”). If the price closes below the start of wave 1 (−3.6%), the setup is void – without an entry of the rule.
Exit with a gain
Buy zone – Rule trigger: candle close above the swing high
Exit at a loss
Watcher Target 1, Target 2 and stop for this zone – with distance in percent and reward/risk – are available from Watcher. Plans and pricing Percentages from the top of the buy zone (61.8%) – not a forecast.
Rule-based, long only, 4-hour candles. Not a forecast and not investment advice. Lesson: the buy zone in 4 minutes
American Express · Sector Financials · amounts in USD
Information only – does not affect our buy zones, not investment advice. Source: SEC filings, As of Feb 6, 2026.
Summarized from the lights below. Describes the reported figures – not a recommendation to buy or sell.
Fiscal year 2025 (Jan 1, 2025 – Dec 31, 2025): Revenue 41.3B USD · Net income 10.8B USD · Net margin 26.2%
Each light follows a simple, disclosed rule. The comparison is with the sector median Sector medianThe middle value of all companies in a sector: half are above it, half below. Unlike the average, single outliers hardly distort it.Example: Margins 4%, 8%, 10%, 12%, 60% → median 10%, average 18.8%.More in the Academy: Lesson putting the P/E in context of all 79 symbols in the Financials sector on our list (current members of the S&P 500 and other lists).
+13.5% per year, last 5 years
Sector Financials: median +8.8% (73 symbols)
What does this mean? Revenue: over 5 years it grew by 13.5% per year on average – at least as fast as the median company in the sector (+8.8%).
Light rule: Green: grows at least as fast as the sector median (without a sector comparison: grows) · Yellow: grows more slowly · Red: shrinks.
+28.1% per year, last 5 years
Sector Financials: median +13.8% (71 symbols)
per share: +32.5% per year (5 years, split-adjusted)
What does this mean? Total net income: over 5 years it grew by 28.1% per year on average – at least as fast as the median company in the sector (+13.8%).
Light rule: Green: grows at least as fast as the sector median (without a sector comparison: grows) · Yellow: grows more slowly · Red: shrinks.
26.2% last fiscal year
Sector Financials: median 19.4% (75 symbols)
What does this mean? Of every 100 USD in revenue, 26.23 USD remained as profit – at least as much as the median company in the sector (19.4%). Five years ago the margin was 14.3%.
Light rule: Green: at least the sector median (without a sector comparison: profit) · Yellow: profit, but below the median · Red: loss.
1.64 latest balance sheet
Sector Financials: median 0.63 (47 symbols)
What does this mean? For every 1 USD of equity there is 1.64 USD of financial debt – more than twice as much as the median company in the sector (0.63). For banks and insurers, debt is part of the business – so the comparison is only with their own sector.
Light rule: Green: at most the sector median · Yellow: up to twice the median · Red: more than twice · Gray: without a sector comparison.
21% Dividend / earnings per share
Sector Financials: median 34% (58 symbols)
What does this mean? Of every 1 USD of earnings per share, 0.21 USD went to shareholders as a dividend. The rest stays in the company. Median company in the sector: 34%.
Light rule: Green: up to 60% of earnings · Yellow: up to 100% · Red: more than earnings or a dividend despite a loss · Gray for real estate (REITs). The sector median is for comparison only.
10 of 10 fiscal years with a profit
What does this mean? In 10 of the last 10 fiscal years the bottom line showed a profit. There were no losses in this period.
Light rule: Green: profit in every year · Yellow: one loss year · Red: several loss years (at least 3 years needed).
| Year | End | Revenue | Net income | Margin | Earnings per share | Dividend per share | Payout |
|---|---|---|---|---|---|---|---|
| 2025 | Dec 31, 2025 | 41.3B | 10.8B | 26.2% | 15.38 | 3.28 | 21% |
| 2024 | Dec 31, 2024 | 38.8B | 10.1B | 26.1% | 14.01 | 2.80 | 20% |
| 2023 | Dec 31, 2023 | 37.2B | 8.4B | 22.5% | 11.21 | 2.40 | 21% |
| 2022 | Dec 31, 2022 | 34.2B | 7.5B | 22.0% | 9.85 | 2.08 | 21% |
| 2021 | Dec 31, 2021 | 27.7B | 8.1B | 29.1% | 10.02 | 1.72 | 17% |
| 2020 | Dec 31, 2020 | 22.0B | 3.1B | 14.3% | 3.77 | 1.72 | 46% |
| 2019 | Dec 31, 2019 | 28.2B | 6.8B | 24.0% | 7.99 | 1.64 | 21% |
| 2018 | Dec 31, 2018 | 26.6B | 6.9B | 26.0% | 7.91 | 1.48 | 19% |
| 2017 | Dec 31, 2017 | 24.8B | 2.7B | 11.0% | 2.97 | 1.34 | 45% |
| 2016 | Dec 31, 2016 | 24.2B | 5.4B | 22.3% | 5.65 | 1.22 | 22% |
| Quarter | End | Revenue | Net income | Margin | Earnings per share | Dividend per share | Payout |
|---|---|---|---|---|---|---|---|
| Q3/26 | Sep 30, 2026 | 11.2B | 3.7B | 33.0% | 5.42 | 0.95 | 18% |
| Q2/26 | Jun 30, 2026 | 11.2B | 3.1B | 27.7% | 4.53 | 0.95 | 21% |
| Q1/26 | Mar 31, 2026 | 10.5B | 3.0B | 28.2% | 4.28 | 0.95 | 22% |
| Q4/25* | Dec 31, 2025 | 10.9B | 2.5B | 22.5% | – | 0.82 | – |
| Q3/25 | Sep 30, 2025 | 10.4B | 2.9B | 27.9% | 4.14 | 0.82 | 20% |
| Q2/25 | Jun 30, 2025 | 10.3B | 2.9B | 28.0% | 4.08 | 0.82 | 20% |
| Q1/25 | Mar 31, 2025 | 9.6B | 2.6B | 26.8% | 3.64 | 0.82 | 23% |
| Q4/24* | Dec 31, 2024 | 10.0B | 2.2B | 21.8% | – | 0.70 | – |
| Q3/24 | Sep 30, 2024 | 9.7B | 2.5B | 25.8% | 3.49 | 0.70 | 20% |
| Q2/24 | Jun 30, 2024 | 9.8B | 3.0B | 30.7% | 4.15 | 0.70 | 17% |
| Q1/24 | Mar 31, 2024 | 9.3B | 2.4B | 26.1% | 3.33 | 0.70 | 21% |
| Q4/23* | Dec 31, 2023 | 9.6B | 1.9B | 20.2% | – | 0.60 | – |
* Q4 derived: fiscal year minus the 9-month value (EPS cannot be derived this way); may be distorted by restatements.
How we calculate: values as first reported to the SEC (later corrections are left out), a fixed order of reporting tags per figure, partial sums are never added up. Growth per year Growth per year (CAGR)The average annual growth rate over several years, as if the value had grown equally strongly every year. Individual good or bad years even out in it.Example: Revenue €100m → €161m in 5 years → (161 ÷ 100) to the power of 1/5 − 1 = 10% per year.More in the Academy: Lesson debt, dividends and growth over 5 years (otherwise 3 years). Quarters are labeled by the calendar quarter in which they end.
| Period | Purchases | Sales | Net |
|---|---|---|---|
| 3 months | $00 people | $3.0M1 person | −$3.0M |
| 6 months | $00 people | $5.4M2 people | −$5.4M |
| 12 months | $00 people | $60.3M10 people | −$60.3M |
Source: mandatory filings with the SEC (Form 4), open-market purchases and sales only. General information, not investment advice.
ATR ATR (Average True Range)The average range per trading day over the last 14 days. It makes distances comparable, whether a stock costs 10 or 1,000.Example: ATR 2.50 at a price of 100: 0.4 ATR is 1.00, i.e. 1%.More in the Academy: Lesson ATR and moving averages as price values only when logged in.
Latest news
No posts on AXP yet – log in to join the discussion.
Posts are opinions of users and only visible after logging in. Community rules