Two companies can sell the same thing to the same buyers and still be very different investments. When a shop pays to show you a pair of sneakers, the money can go to a social app you scroll at night or to the online store you buy from. Both firms count it as ad revenue.
The trouble comes when you try to compare the two shares. One company earns nearly all its money that way, while the other earns most of its sales from shipping goods and renting out computers. A headline figure such as total profit or share price growth hides more than it shows.
Investors handle this by laying the two businesses side by side, line by line. They check where each dollar of sales comes from and how much of it ends up as profit. Then they look at what each spends on data centers and what each hands back to owners, over the same months.
Meta Platforms owns Facebook, Instagram and WhatsApp, while Amazon runs a global store and Amazon Web Services (AWS). Both sell ads, and both are spending record sums on AI. BloFin gives you two ways to trade Meta. The METAX token, which you buy and hold on the METAX/USDT Spot market, tracks the share price, while the METAUSDT Perpetual is a leveraged contract that gains when that price moves your way, up or down.
What is the difference between Meta and Amazon stock?
Meta and Amazon stock differ mainly in how much each company depends on ads, a business they both run. From April to June 2026, advertising made up 97.6% of Meta's revenue and 9.9% of Amazon's (source: Meta Q2 2026 results) (source: Amazon Q2 2026 results).
So a Meta share is close to a pure bet on one product: ads shown to the 3.6 billion people a day who used its apps on average in June 2026 (source: Meta Q2 2026 results). An Amazon share holds three businesses at once: its stores bring in most of the sales, and AWS rents computing power to other firms. Ads are a smaller line that is growing fast, up 26.2% on a year earlier (source: Amazon Q2 2026 results).
That one difference drives most of the other contrasts between the two shares. When ad budgets grow, Meta feels it across almost all of its sales, while Amazon feels it on about a tenth. When firms rush to rent AI computing power, Amazon sells it to them, and Meta is one of the buyers. Meta's advertising business is the whole engine on one side, while on the other ads are one part of a much larger machine.
The way each company reports its results shows the same split. Meta reports two segments: its Family of Apps, and Reality Labs, the unit that makes its headsets and glasses (source: Meta 10-Q for the second quarter of 2026). Amazon reports three, which are North America, International and AWS (source: Amazon Q2 2026 results). Owning a share of Meta stock or Amazon stock means owning a small slice of every one of those segments.
Revenue, growth and margins in the June 2026 quarter
In the April to June quarter of 2026, Meta's revenue grew 28.0% to $60.8 billion and Amazon's net sales grew 19.6% to $200.6 billion. Amazon sold about 3.3 times as much, but Meta kept far more of each dollar as profit: its operating margin was 30.9%, against 13.7% at Amazon (source: Meta Q2 2026 results) (source: Amazon Q2 2026 results).
Ads brought Meta $59.4 billion in the quarter and Amazon $19.8 billion, while Meta's operating income was $18.8 billion and Amazon's was $27.5 billion (source: Meta Q2 2026 results) (source: Amazon Q2 2026 results).
April to June 2026 | Meta | Amazon |
|---|---|---|
Revenue | $60.8 billion | $200.6 billion |
Growth on a year earlier | 28.0% | 19.6% |
Advertising revenue | $59.4 billion | $19.8 billion |
Advertising as a share of revenue | 97.6% | 9.9% |
Operating income | $18.8 billion | $27.5 billion |
Operating margin | 30.9% | 13.7% |
Operating margin a year earlier | 43.0% | 11.4% |
The margins moved in opposite directions over the year. Meta's fell from 43.0% as its costs rose faster than its sales, and those costs included $1.2 billion of severance pay (source: Meta 10-Q for the second quarter of 2026). Amazon's rose from 11.4%, and AWS supplied about 78% of the $8.3 billion rise in its operating income (source: Amazon Q2 2026 results).
You get operating margin by dividing operating income by sales, and the answer is the number of cents from every dollar of sales that remain after the business pays its running costs. At Meta that was about 31 cents in the June quarter, and at Amazon about 14 cents. Even so, Amazon's operating income was about 1.5 times Meta's, because it sells so much more.
Both margins also depend on a choice about how long servers last. From January 1, 2025, Meta stretched the expected life of most servers and network gear to 5.5 years, which spreads their cost over more years (source: Meta 2025 annual report). Amazon went the other way on the same date, cutting the life of some of its servers from six years to five. It said technology, and AI above all, was now changing faster (source: Amazon 2025 annual report). A longer life lowers the yearly cost charged for wear, called depreciation, so Meta's choice spreads the cost of its AI capex over more years and lifts the profit it reports.
Market value and valuation at the October 20 close
At the October 20, 2026 close, Amazon was worth $2.8 trillion and Meta $1.8 trillion, so Amazon was valued at about 1.6 times Meta. Meta traded at 26.2 times its past year's earnings and 21.8 times forecast earnings, while Amazon traded at 21.1 and 27.9 (source: Stock Analysis Meta statistics) (source: Stock Analysis Amazon statistics).
At the October 20, 2026 close | Meta | Amazon |
|---|---|---|
Market value | $1.8 trillion | $2.8 trillion |
Price to past year's earnings (trailing P/E) | 26.2 | 21.1 |
Price to forecast earnings (forward P/E) | 21.8 | 27.9 |
Price to sales | 7.75 | 3.65 |
The two P/E ratios point in opposite directions, and one line in Amazon's results explains much of it. Its second-quarter net income of $62.6 billion included $53.4 billion of pre-tax other income, "primarily from our investments in Anthropic" (source: Amazon Q2 2026 results). Anthropic is an AI lab Amazon has backed since 2023. Gains on Amazon's Anthropic stake lift its past-year earnings and make its trailing P/E look low, while forecasts of future profit usually leave them out.
Price to sales is the plainer comparison, because sales leave out the one-off gains and charges that move Meta's P/E ratio and Amazon's. Investors paid about $7.75 for each dollar of Meta's yearly sales and about $3.65 for each dollar of Amazon's. The gap mostly reflects the margins, since a dollar of Meta's sales carries far more profit than a dollar of store sales.
Share price returns and the largest falls since 2021
To the October 20, 2026 close, Amazon's shares had gained 13.8% since the start of the year and 21.3% over twelve months, against a 5.2% gain and a 5.2% loss for Meta's. Across five years the order flips: Meta rose 103.7% and Amazon 53.8% (source: Yahoo Finance META history) (source: Yahoo Finance AMZN history).
Return to the October 20, 2026 close | Meta | Amazon |
|---|---|---|
Since December 31, 2025 | +5.2% | +13.8% |
One year, from October 20, 2025 | −5.2% | +21.3% |
Five years, from October 20, 2021 | +103.7% | +53.8% |
Deepest peak-to-trough fall in those five years | −74.4% (source: Yahoo Finance META history) | −55.7% (source: Yahoo Finance AMZN history) |
These are share price changes only, with Amazon's prices adjusted for its 20-for-1 stock split in June 2022. Meta's dividends would lift its figures slightly for someone who received them, and Amazon has none. Measured that way, a $10,000 stake bought in Meta at the close on October 20, 2021 was worth roughly $20,372 by October 20, 2026, against roughly $15,381 for the same sum in Amazon.
The bigger five-year gain came with the deeper fall. Meta's close dropped 74.4% from November 15, 2021 to November 3, 2022, while Amazon's fell 55.7% from November 18, 2021 to December 28, 2022. Both slides came in the same 2022 sell-off in tech shares, yet Meta fell further and then climbed further. That fall cost Meta's holders close to three quarters of their stake's value in under a year, a risk worth setting against the five-year gain when you judge whether Meta is a good investment.
Start dates matter as much as the companies here. Over the past year Amazon led, helped by a 15.3% jump on July 31, 2026, the day after its second-quarter report. Meta fell 3.7% on October 16 alone (source: Yahoo Finance AMZN history) (source: Yahoo Finance META history). A window that starts or ends near one sharp day can tell a very different story, and past returns say nothing about which share does better next.
Dividends, buybacks and share counts
Meta pays $0.525 a share in dividends each quarter and Amazon pays none. Neither company bought back any of its own shares from January to June 2026, so both share counts rose. Meta's grew 0.7% to 2,548 million and Amazon's 0.5% to 10,783 million (source: Meta 10-Q for the second quarter of 2026) (source: Amazon 10-Q for the second quarter of 2026).
That is a change for Meta, since a year earlier, from January to June 2025, its count fell by 0.7% while it was buying back stock. It still had $25.0 billion of approved buybacks left unused at June 30, 2026 (source: Meta 10-Q for the second quarter of 2026). Amazon's buyback program had $6.1 billion left on the same date (source: Amazon 10-Q for the second quarter of 2026). Meta also paid $2.7 billion in dividends to its shareholders in the first half of 2026, while it spent heavily on data centers.
A rising share count works against each owner in a small way. Shares handed to staff as pay are added to the total, so each existing share stands for a slightly smaller piece of the company until buybacks take them out again. In the first half of 2026 that effect, called dilution, was similar at the two companies, at well under 1% each.
Say you hold 10 Meta shares for a year at the current dividend. You would receive $21.00 before tax, which is $2.10 a share, a yield of about 0.30% at the October 20 close of $694.25. Ten Amazon shares would pay you nothing, so any return there comes only from the share price.
AWS and the computing power Meta rents
AWS is the main reason the two companies earn money differently. In the June 2026 quarter AWS made $42.2 billion of sales, up 37%. That was 21.1% of Amazon's sales but 60.5% of its operating income (source: Amazon Q2 2026 results). Meta's two segments earn money from ads and devices, while AWS earns it by renting computers.
Amazon sells its AI build to others through AWS. In the same results it said its chips business and the AI business inside AWS had each passed a $25 billion yearly run rate. It also said Anthropic and OpenAI had made "multi-year, multi-gigawatt commitments" to Trainium, the AI chip Amazon designs for its own data centers (source: Amazon Q2 2026 results).
A run rate multiplies the most recent period's sales out to a whole year, so it shows how fast a business is selling right now. AWS's own run rate was $169 billion, and its operating margin in the quarter was 39.4%, above the 30.9% Meta earned across its whole company. That is why AWS carries most of Amazon's profit while bringing in about a fifth of its sales.
Meta sits on the other side of that trade. It builds and fills its own data centers, but it also signs compute deals to rent AI capacity from outside providers. At June 30, 2026 it had $349.3 billion of contracts it cannot cancel, mostly for "third-party cloud capacity arrangements" and servers, network gear and data centers (source: Meta 10-Q for the second quarter of 2026).
Amazon's Muse block and the deals between Meta and Amazon
Amazon began shutting Meta's Muse AI agent out of its online store on the night of Sunday, September 20, 2026, according to a company spokesperson. Amazon had also asked Meta to take Amazon out of the agent (source: Bloomberg via The Star). The fight matters more because the two also buy from each other, in cloud computing and in ads.
The spokesperson, Lara Hendrickson, said shopping apps acting for a customer should "operate openly and respect service provider decisions about whether or not to participate". She added that AI agents such as Muse are bound by the same duty (source: Bloomberg via The Star). An AI agent is a program that acts on your behalf, for example by finding a product and paying for it. Meta's Muse AI shops this way, and Meta has spoken about how it may earn money from it.
The block touches both shares through the same shopper. Amazon made $19.8 billion from ads in the June quarter (source: Amazon Q2 2026 results). Bloomberg noted that shopping bots could steer buyers to other stores, which would cost Amazon sales and ad revenue (source: Bloomberg via The Star). Muse is built to shop for you, so each store that shuts it out makes it a little less useful to the people Meta hopes will use it.
Five months earlier the two had signed a deal in the other direction. On April 24, 2026, Amazon said Meta would use tens of millions of cores of AWS's Graviton chips to run AI agents, and it called Meta "a longtime AWS customer" (source: Amazon on the Meta Graviton agreement). Meta's infrastructure head said diversifying its compute sources is "a strategic imperative". Part of what Meta spends on AI therefore turns into AWS revenue.
The ad link has also run both ways. In November 2023 the two launched a feature that let US shoppers who linked a Facebook or Instagram account to Amazon buy from select Amazon product ads without leaving Meta's apps. An Amazon spokesperson confirmed the launch at the time (source: TechCrunch). Amazon has bought ads on Meta's apps and Meta pays Amazon for computing power, yet in September 2026 the two fell out over who controls the shopper.
AI spending and how each company pays for it
From January to June 2026, Meta spent 41.9% of its sales on property and equipment and Amazon spent 25.8%. Against the cash their operations brought in, Meta spent 76.6% of it. Amazon spent 137.8%, which is more than its operations produced (source: Meta 10-Q for the second quarter of 2026) (source: Amazon 10-Q for the second quarter of 2026).
Capex, short for capital expenditure, is money paid for assets used over many years, like data centers, servers and warehouses. Amazon spends more dollars: $98.4 billion in those six months, against Meta's $49.1 billion. Meta spends more of each sales dollar, and it expects $130 billion to $145 billion of capex for all of 2026 (source: Meta Q2 2026 results).
The cash test is where the two part ways, and free cash flow is the cash left after capex. Meta's was $784 million in the June quarter alone, a thin amount after a large build. Amazon gives its figure for a full year, and in the year to June 30, 2026 it was minus $7.6 billion, so Amazon spent more on its build than its operations brought in (source: Amazon Q2 2026 results).
Both are borrowing to help pay for it. The face value of Amazon's long-term debt rose from $68.8 billion at December 31, 2025 to $133.0 billion at June 30, 2026, as it sold notes in four currencies. In June it also signed a $17.5 billion term loan, a bank loan it could draw in one go until September 30, 2026. It had borrowed nothing under it at June 30 (source: Amazon 10-Q for the second quarter of 2026).
Meta carried $83.7 billion of long-term debt at June 30, 2026 (source: Meta Q2 2026 results). It also pays for part of its build through a venture to develop a data center campus in Louisiana. Meta holds a 20% stake in that venture and does not consolidate it on its own balance sheet (source: Meta 10-Q for the second quarter of 2026). That venture and Meta's bond sales are the core of how Meta is financing the AI build.
The third-quarter reports land eight days apart. Meta set its third-quarter 2026 results for after the US market close on Wednesday, October 21, 2026 (source: Meta third-quarter 2026 results notice), and Amazon set its third-quarter call for Thursday, October 29, 2026, at 5PM ET (9PM UTC) (source: Amazon third-quarter 2026 call notice). Each report shows whether the spending, and the borrowing behind it, is still rising.
Amazon trades on BloFin beside Meta, both as a token on the AMZNX/USDT Spot market and as the AMZNUSDT Perpetual contract.
METAX and the METAUSDT Perpetual move with Meta's shares on nights like these. If you want to watch how the token reacts to both reports, the METAX/USDT Spot page shows its price as it trades, and the METAUSDT Perpetual page adds the contract's funding rate. That rate is a small fee one side pays the other at set times, and it pulls the contract back toward the share price. Outside the Nasdaq session, BloFin may accept only orders that reduce a position on the perpetual, so on a report night you may only be able to close or shrink one until the next session opens (source: BloFin stock futures notice).
Antitrust and legal risks for Amazon and Meta
Amazon faces a federal antitrust trial, brought by the Federal Trade Commission (FTC) and a group of states, set to begin on March 29, 2027. In October 2026, US District Judge John Chun refused Amazon's request to move it to June 2027. He wrote that he did not "see a good reason to disturb the current schedule" (source: MLex).
The FTC, state attorneys general and private plaintiffs have filed a group of cases against Amazon since March 2020, its filings say. They accuse it of fixing prices with its sellers and of monopolization, and Amazon says it disputes the claims and will defend itself (source: Amazon 10-Q for the second quarter of 2026).
Meta's biggest antitrust case has already gone to trial. The FTC tried to make it sell Instagram and WhatsApp, a judge ruled for Meta on November 18, 2025, and the FTC is appealing (source: Meta 10-Q for the second quarter of 2026). While the appeal runs, the FTC antitrust case and Meta stock stay tied together.
The two legal risks reach different parts of each company, since Amazon's trial is about how its store treats sellers and shoppers. If the FTC wins its appeal, that would reopen the question of whether Meta can keep the two apps it bought in 2012 and 2014.
How to choose between Meta and Amazon stock
Choosing between Meta and Amazon stock comes down to which business you want your money tied to. Meta is a concentrated bet on advertising, with 97.6% of its revenue from ads and an operating margin of 30.9% in the June 2026 quarter (source: Meta Q2 2026 results). Amazon spreads you across its stores, its ads and AWS, which supplied 60.5% of its operating income on 21.1% of its sales (source: Amazon Q2 2026 results).
If you want your exposure tied to ad budgets and can live with sharper falls, Meta's record shows the trade-off: its deepest drop in the five years to October 2026 was 74.4%, against 55.7% for Amazon (source: Yahoo Finance META history) (source: Yahoo Finance AMZN history). If you would rather own a seller of AI computing power, Amazon earns from that through AWS while Meta pays for it. Meta pays a small dividend and Amazon pays none, and both are borrowing to fund their data centers. These figures describe the two businesses, and they say nothing about which share will do better from here; BloFin does not publish price targets or forecasts.
How Meta compares with other big-tech stocks and Bitcoin
Amazon is one of several big-tech shares Meta is often weighed against, and each pairing turns on a different question. Alphabet is the other giant of online ads, Microsoft sells AI to businesses through its cloud and Nvidia makes the chips Meta buys. Bitcoin tests whether a crypto holder already owns Meta's risk.
Alphabet is the closest match, because Alphabet and Meta sell the same kind of product to advertisers, Google beside search results and YouTube videos and Meta inside its own feeds.
Microsoft and Nvidia sit on the selling side of AI, like AWS. Microsoft rents computing power and AI models to businesses through Azure, while Meta gives its apps away and is paid by advertisers, the main contrast between Meta and Microsoft stock. Nvidia earns much of its revenue from companies building AI data centers, Meta among them, so Meta and Nvidia stock put the buyer of the chips beside the seller.
If you already hold crypto, the pairing that matters most may be the one with Bitcoin, because the correlation between Meta stock and Bitcoin changes with the months you measure.
Looking to trade Meta or Amazon? To get started, you'll need to first create a BloFin account, fund your account with cryptocurrency, and go to the METAX/USDT Spot trading page, the METAUSDT Perpetual trading page, the AMZNX/USDT Spot trading page or the AMZNUSDT Perpetual trading page.
Frequently asked questions
How many employees do Meta and Amazon have?
Amazon had 1,595,000 full-time and part-time employees at June 30, 2026, not counting contractors and temporary staff (source: Amazon Q2 2026 results). Meta had 75,472, down 1% on a year earlier (source: Meta Q2 2026 results). Amazon therefore employs about 21 times as many people, while in the June 2026 quarter Meta brought in about $805,600 of revenue per employee against about $125,800 at Amazon.
Which company made more profit, Meta or Amazon?
In the June 2026 quarter, Meta's net income was $15.8 billion, 14% less than a year earlier (source: Meta Q2 2026 results). Amazon's was $62.6 billion, about 3.4 times its $18.2 billion of a year earlier, and that figure included its gains on Anthropic (source: Amazon Q2 2026 results). Net income counts those gains, so it gives Amazon a far bigger lead than operating income does.
Does Meta hold stakes in private companies, as Amazon does in Anthropic?
Yes. Meta carried its investments in privately held companies at $30.2 billion at June 30, 2026, including its 20% stake in the Louisiana data center venture (source: Meta 10-Q for the second quarter of 2026). It has written up the stakes it carries at cost by only $559 million in total, a small sum next to the $53.4 billion of other income Amazon booked in one quarter, mostly from Anthropic.
How long did Meta and Amazon stock take to recover from 2022?
Meta's close first got back above its November 15, 2021 level on December 19, 2023, about 13.5 months after its November 2022 low. Amazon's close regained its November 18, 2021 level on April 5, 2024, about 15 months after its December 2022 low (source: Yahoo Finance META history) (source: Yahoo Finance AMZN history).
Which stock is more volatile, Meta or Amazon?
Over five years Amazon's beta was 1.46 and Meta's was 1.19, at the October 20, 2026 close (source: Stock Analysis Amazon statistics) (source: Stock Analysis Meta statistics). Beta compares a share's past swings with those of the whole market, so by that measure Amazon swung more. Meta still had the deeper single fall, so beta and the worst drop can tell different stories.
Researched and written by the BloFin Academy editorial team with AI-assisted drafting. All facts independently verified. Primary sources include Meta's and Amazon's second-quarter 2026 results and 10-Q filings, both companies' 2025 annual reports, Amazon's Graviton announcement, Meta's and Amazon's third-quarter results notices, Stock Analysis and Yahoo Finance price history, and BloFin's stock futures notice. News sources include Bloomberg, TechCrunch and MLex, current as of October 21, 2026.
Nothing in this article constitutes financial advice, and nothing in it is a recommendation to buy, sell or hold Meta or Amazon in any form. Meta's shares fell 74.4% from a peak within the past five years and Amazon's 55.7%, and either can drop sharply on a results night or a court ruling. METAX follows Meta's share price, and a leveraged METAUSDT Perpetual position can lose its whole margin in a single gap. Past performance does not indicate future results. Do your own research and consider your risk tolerance before you trade on BloFin.