How to Trade Amazon (AMZN): Hours, Earnings and What Moves It

Amazon looks like a retailer and trades like a cloud computing company. Most of its revenue comes from selling things; most of its profit comes from renting out servers, and it is the second of those that decides which way the stock gaps on earnings night.

Amazon (AMZN) at a glance

What you are actually trading

An Amazon CFD on MT5 is a contract with your broker that settles the difference between your opening and closing price. There is no share, no shareholder vote and no claim on the company. What you gain is leverage, the ability to short as easily as you go long, and precise sizing. What you pay, beyond the spread, is overnight financing calculated on the full notional value of the position rather than on your margin, which makes share CFDs suitable for days and weeks, not for quarters.

What you are trading is a bet on two cycles at once. The retail business is exposed to the consumer: employment, wages, credit conditions, fuel and freight costs, and the strength of the holiday season. The cloud business is exposed to corporate technology budgets and, increasingly, to AI infrastructure demand. Those two cycles do not always point the same way, which is why Amazon can report record sales into a soft stock reaction, or weak retail into a rally because AWS reaccelerated.

Amazon has also earned a reputation for large earnings moves. Its cost base is enormous and partly fixed, so a small change in revenue or in how much it is spending on logistics and data centres produces a large change in operating profit. That operating leverage is the mathematical reason the stock gaps hard: the market is not adjusting to a 2% revenue surprise, it is adjusting to what that 2% did to profits and to next quarter’s guidance range.

Finally, you are trading an index heavyweight. Amazon is one of the largest weights in the S&P 500 and the Nasdaq-100, so passive flows buy and sell it mechanically, and a large AMZN move pushes the index rather than merely following it. On any ordinary day with no Amazon news, most of the stock’s move is simply the market moving.

What moves the price

AWS growth rate and margin

This is the single most important input. AWS is the profit engine, so the market watches its year-on-year growth rate and its operating margin far more closely than anything in the retail business. An acceleration in that growth rate has repeatedly lifted the stock even when retail was unremarkable; a deceleration has sunk it on an otherwise strong quarter.

It is also a competitive number. Microsoft’s Azure figures and Alphabet’s cloud figures arrive within days, so the market judges AWS in relative terms, losing share to a faster-growing rival is punished even if the absolute number looks fine.

Operating income guidance

Amazon guides to a range for next quarter’s operating income, and that range moves the stock as much as the results do. Because the company’s costs are so large, small changes in shipping, warehousing, headcount or data-centre spending produce disproportionate swings in that number. A conservative guide can turn a good quarter into a sharply lower open.

The consumer and the holiday quarter

Retail demand tracks the health of the consumer: employment reports, retail sales data, consumer confidence and credit conditions all read across. The October to December quarter is by far the largest, and it is reported in the first results of the following year, so the January or early February report carries more weight than the others. Prime Day in the summer and the peak shipping season are the intermediate checkpoints the market uses.

Capital expenditure and AI infrastructure

Amazon spends heavily on warehouses, delivery and now on AI data centres. The market’s interpretation of that spending swings between reading it as evidence of demand and reading it as margin destruction. The same capital-expenditure figure can lift or sink the stock depending on which mood prevails, and it simultaneously moves suppliers such as NVIDIA.

Interest rates and the market

Amazon’s valuation leans on future profits, so higher interest rates reduce what those profits are worth today. CPI at 08:30 New York time and Fed decisions at 14:00 therefore move AMZN, usually by moving the whole index. Rates also hit the retail side directly, because consumer credit becomes more expensive. Beta, how far a stock tends to move for a given index move, sits above 1 here, so Amazon typically exaggerates the market’s direction.

Regulatory and labour risk

Antitrust action over marketplace practices, scrutiny of the bundling of Prime services, and unionisation efforts across its warehouse network all arrive on legal and political timetables rather than financial ones. They rarely produce the largest moves, but they produce unscheduled ones, which is an argument for keeping overnight positions sized for a surprise.

The best time of day to trade Amazon (AMZN)

The cash session runs 09:30 to 16:00 New York time: 14:30 to 21:00 UK time for most of the year, with a short period of drift around the daylight-saving changeovers. Nearly all real volume happens there, and so do the tightest spreads. The New York session guide puts the window in context.

Pre-market runs from 04:00 New York time to the open, after-hours from the close to 20:00. Both print prices and both are thin: the order book is shallow, spreads are wide, and a modest order can shift the price several percent. Amazon releases earnings into the after-hours window, followed by a call, which is why the number you see on the news at 16:30 is set by very little money and frequently changes shape by the following morning.

Most CFD brokers quote AMZN only during or just around cash hours, so a position you hold overnight is frozen while the news that will reprice it is being digested elsewhere. That is not a broker failing; it is the structure of the market you have chosen to trade.

Times follow the live session clock. Use the forex market hours tool to convert any of these into your own timezone, and see the session times hub for why fixed UTC tables are wrong half the year.

How different traders approach it

If you are brand new

Amazon is a good stock to learn on in normal conditions: deeply liquid, tight spreads in the cash session, and clean enough trends to practise on. The risk is concentrated into four nights a year, and it is a big risk.

Be flat into earnings. Amazon has a long history of double-digit percentage moves after reporting, in both directions. The report comes after the closing bell, roughly in late January or early February, late April, late July or early August, and late October. The stock then reopens wherever it reopens. A stop-loss does not protect you: a stop tells your broker to trade at the next available price once a level is touched, and if the next available price is 12% away, that is your fill. There is no safe version of a normal-sized overnight position through this report.

Trade the cash session. 09:30 to 16:00 New York time, and preferably 10:30 to 11:30 while you are learning. Outside that, liquidity is poor and the price is unreliable.

Check the index and the sector. Look at the Nasdaq-100 before you take a trade, on an ordinary day most of Amazon’s move is the market. Risk 0.5% to 1% per trade and use the position size calculator rather than guessing a size.

If you already trade but results are inconsistent

The intermediate mistake specific to Amazon is trading the retail story. You read about record holiday sales or a strong Prime Day, form a bullish view, and then watch the stock fall on a cloud growth rate you never checked. Revenue lives in retail; profit lives in AWS. If your thesis does not mention AWS growth and the operating income guidance range, it is not a thesis about what moves this stock.

The second is holding a swing position across the report because the technical setup is too good to abandon. Amazon’s operating leverage means the gap distribution is genuinely wide, so this is not a case of accepting a slightly larger risk; it is a case of your risk plan not applying. If you want earnings exposure, size for the gap: work out what a 12% adverse move costs, and make that your risk figure, which usually means a fraction of your normal position.

Third, use the cloud read-across. Microsoft and Alphabet report their cloud numbers within days of Amazon, and a strong or weak print from either reprices expectations for AWS before Amazon has said anything. That read-across is a tradeable, repeatable pattern and most retail traders ignore it entirely.

If you are experienced

Amazon is an operating-leverage story wearing a retailer’s clothes. The fixed-cost base across fulfilment and infrastructure means small revenue and cost surprises translate into large swings in operating income, which is why the implied move priced ahead of the report is consistently among the widest in the mega-cap complex and why realised moves have so often justified it. Volatility sellers get punished here more than on Microsoft or Apple.

The cloud complex trades as a group with a staggered calendar. Microsoft, Alphabet and Amazon report within a narrow window, so the first print resets expectations for the others, and the correlation of the reaction is unstable; a strong Azure figure can be read as either sector strength or share loss for AWS, and which reading prevails depends on the prevailing narrative rather than on the numbers. Watching how the second reporter trades on the first reporter’s call is the cleanest available signal about which regime is in force.

Capital expenditure is the regime variable to track. The market alternates between rewarding AI infrastructure spend as demand confirmation and punishing it as margin destruction, and the flip typically shows first in the divergence between the hyperscaler and its suppliers on the same announcement. Index weight adds a second layer: a large AMZN gap propagates into the S&P 500 and Nasdaq-100 directly, so single-name and index exposure are less independent than they appear.

Strategies that work on Amazon (AMZN)

Trend continuation on pullbacks : swing traders, beginners upwards

Amazon trends reasonably cleanly between catalysts, which suits buying orderly pullbacks in an uptrend and selling rallies in a downtrend rather than chasing breakouts. Establish the trend on the daily chart, wait for a retracement into a prior structural level or a well-respected moving average, and enter when the daily bars stop extending against you.

Two constraints. Check the earnings date before entering, if it falls inside your holding period, wait or cut the size sharply. And remember that financing is charged nightly on the full notional value, so a slow trade has a running cost.

Post-earnings range break : everyone; four dates a year

Be flat before the close on earnings day. Let the gap happen, then trade the session that follows, typically one of the highest-volume, cleanest-trending days of the quarter.

Let the first 15 to 30 minutes of the cash session build a range in the repriced stock and trade the break of that range. Levels from before the gap carry much less weight, because the market has genuinely new information about AWS and about guidance. If the gap starts filling back towards the previous close inside the first hour, that failure is often the better trade, and it tends to run.

Cloud read-across from Microsoft and Alphabet : advanced; a handful of dates a year

Microsoft and Alphabet report cloud growth within days of Amazon. A clear acceleration or deceleration from either reprices expectations for AWS immediately, before Amazon has reported anything itself.

Trade the read-across in the cash session following the other company’s report, not in after-hours. Be aware the sign can invert: a rival’s strong cloud number can be read as sector demand, which lifts AMZN, or as share loss, which sinks it. Let the first half hour of real liquidity tell you which interpretation the market has chosen before committing.

Opening range with an index filter : intraday traders

Mark the high and low of the first 15 or 30 minutes of the cash session, then take the break only in the direction the Nasdaq-100 is also breaking. Amazon’s beta is above 1, so when it and the index agree the move has real flow behind it; when they disagree, the index usually wins.

Stop the other side of the opening range, target a multiple of its height, and stop trading after 11:30 New York time.

Common mistakes on Amazon (AMZN)

• Trading the retail headline instead of the cloud number. Most of Amazon’s revenue is retail, but most of its profit, and nearly all of the market’s attention, is AWS.
• Holding through earnings at normal size. Amazon has a long record of double-digit percentage gaps, and no stop can execute inside a gap.
• Believing the after-hours reaction is final. The move printed at 16:30 New York time is set on tiny volume and often changes materially once the cash session opens.
• Ignoring the operating income guidance. The forward range moves the stock as much as the results, because Amazon’s cost base makes profit far more volatile than revenue.
• Trading the midday lull. The 11:30 to 14:00 New York window produces convincing setups with nobody on the other side of them.
• Assuming Amazon is defensive because it is huge. Its beta is above 1, so it typically falls further than the index on risk-off days, not less.
• Reusing a forex position size. One AMZN CFD carries a large notional value and retail stock CFD leverage is capped far below forex leverage. Recalculate every time.

Risk and position sizing

The specific hazard on Amazon is the earnings gap distribution. This is a company whose profits swing far more than its revenue, so the market’s reaction to a report is genuinely wide-tailed, double-digit percentage moves are not exotic here, they are part of the record. Any overnight position needs to be sized with that in mind rather than sized off the stop.

Work in order. Set the risk percentage per trade, typically 0.5% to 1%. Put the stop where the idea is actually invalidated, using a volatility measure rather than a round number. Convert to contracts with the position size calculator, and do not round up. Then, for anything held overnight, run the gap test: what would a 10% adverse gap cost? If that number is unacceptable, the position is too large no matter where the stop is.

Two further costs. Overnight financing is charged on the full notional value of the position every night it is open, not on your margin, which matters on a strategy built around multi-week holds. And at regulated UK and EU brokers, retail leverage on single-share CFDs is capped at 5:1, a 20% margin requirement, which is far tighter than forex and should be read as a limit rather than a target. If your account is not denominated in US dollars, every result also carries a currency conversion.

Work the numbers before you enter with the position size calculator, the pip value calculator and the risk/reward calculator.

Where Market Structure Pro fits

Amazon’s problem for a discretionary trader is that it produces two quite different markets in the same chart window. Between catalysts it trends smoothly enough to reward patience; around cloud numbers, capex commentary and earnings it becomes a wide, gap-prone, headline-driven instrument. The setups look the same in both regimes, and traders keep applying the same size and the same stop distance to both.

Market Structure Pro is built to make the regime explicit rather than leaving it to feel. It fuses 27 tools into one verdict (TRADE, TRANSITION or NO TRADE) with a confidence percentage, an A/B/C grade and a plain-English account of what is supporting or limiting that verdict. The TRANSITION state is particularly useful on AMZN, because so much of the damage here is done in the phase where a trend has stopped working but has not yet reversed. Its dedicated ranging filter is designed to return NO TRADE in chop, which is most of what the 11:30 to 14:00 window offers.

It is session-aware, so a break in dead midday liquidity is graded for the conditions it is actually in, and spread-aware, which matters on a share CFD where the quote widens the moment volatility rises. State locks on the closed bar, so nothing repaints and your review of the week is honest. MSP is decision support, not a signal service; it does not place trades, it cannot see an earnings release coming, and it guarantees nothing. Being flat into the report remains a decision only you can take.

Frequently asked questions

What actually moves Amazon stock on earnings?

AWS revenue growth and operating margin, together with the guidance range for next quarter’s operating income. The retail business supplies most of the revenue but very little of the profit, so strong sales figures frequently fail to lift the stock if the cloud growth rate has slowed or the guidance is conservative.

When does Amazon report earnings?

Amazon reports on a calendar year, after the closing bell, broadly in late January or early February, late April, late July or early August, and late October. The first report of the year covers the holiday quarter, which is the largest of the four and usually the most closely watched.

Why does Amazon gap so much on earnings?

Its cost base across warehouses, delivery and data centres is enormous and partly fixed, so small changes in revenue or spending produce large swings in operating profit. That operating leverage means the market has to reprice the stock substantially when results or guidance differ from expectations, and double-digit percentage gaps have happened repeatedly.

What are Amazon’s trading hours?

The US cash session runs 09:30 to 16:00 New York time, which is 14:30 to 21:00 UK time for most of the year. Pre-market from 04:00 and after-hours to 20:00 New York time exist but are thin with much wider spreads, and most CFD brokers quote single-share CFDs only during or close to cash hours.

Does a stop-loss protect you against an Amazon gap?

No. A stop is an instruction to trade at the next available price once your level is reached, so if the stock reopens far beyond it you are filled at that new price. Because Amazon has a history of very large earnings gaps, most retail traders should be flat into the report or size the position for a double-digit adverse move.

Does Amazon pay a dividend to CFD traders?

A CFD never pays a company dividend, because you do not own the share and have no voting rights. If a dividend is declared, brokers apply a cash adjustment on the ex-dividend date instead, crediting longs and debiting shorts. Amazon spent its history reinvesting rather than paying dividends, so check your broker’s corporate-action feed for current status.

Is Amazon good for swing trading?

It trends reasonably cleanly between catalysts, which suits multi-day holds, and it is liquid enough that entries and exits are straightforward during cash hours. The two constraints are the earnings calendar, which should be checked before any multi-week position, and overnight financing, which is charged on the full value of the position every night.

How does Amazon compare with Microsoft for cloud exposure?

Both are major cloud providers and their growth rates are compared directly by the market, so a strong number from one reprices expectations for the other within days. Microsoft is the calmer stock with a lower beta and steadier enterprise revenue, while Amazon carries the additional consumer-facing retail cycle and a wider distribution of earnings-day outcomes.

Related instruments

• Microsoft (MSFT): The direct cloud competitor: its Azure figures reprice AWS expectations.
• Alphabet (GOOGL): The third hyperscaler, reporting in the same narrow window.
• Nasdaq 100: Amazon is a top weight; check the index before taking a single-stock view.
• NVIDIA (NVDA): Amazon’s data-centre spending is NVIDIA’s revenue: the two trade off each other.