9%–12% Guided, Prime Day Moved to June (Amazon Earnings Preview, 29 October 2026): What to Expect From Q3 — and Why AWS Has to Clear 36.7%

Amazon reports third-quarter 2026 results after the US close on Thursday 29 October, with the call at 5 p.m. Eastern (21:00 UTC). It has guided net sales of $197bn to $202bn, growth of 9% to 12%, and operating income of $22.5bn to $26.5bn; published consensus puts revenue at about $202bn, the top of the range. The headline growth rate will look like a sharp slowdown from the second quarter's 20%, but most of that is a calendar effect — Prime Day moved into June — and the number the market will actually trade is AWS growth against the 36.7% it posted last quarter.

When the report lands, and what else is in the window

Amazon confirmed in a company announcement that it will hold its third-quarter conference call on Thursday 29 October at 2 p.m. Pacific, 5 p.m. Eastern. US clocks do not change until 1 November, so that is 21:00 UTC. The release is normally posted shortly after the 4 p.m. close, which means the first move in index futures comes from the headline table and a second one often comes during the call, when management talks about capacity and spending.

The report closes out a crowded 36 hours. The Federal Reserve decides on rates and Microsoft and Alphabet report on Wednesday 28 October — we previewed both the Fed decision and Microsoft's quarter. By Thursday evening the market will already have two of the three large cloud providers' growth rates to compare AWS against, plus the Fed's signal on the rate path. That makes AWS a relative read as much as an absolute one.

What the guidance and the consensus imply

Guidance from Amazon's Q2 2026 results; year-ago figures from its Q3 2025 results. Consensus from published analyst tallies, which differ by compiler.

Two points sit inside that table. First, consensus revenue at about $202bn is the top of Amazon's range, so the market is already pricing a result at the upper end — a print in the middle of the guide would be a miss against expectations even though it would meet Amazon's own forecast. Amazon has a habit of guiding low: in the second quarter it reported $200.6bn against a $194bn–$199bn guide, and $27.5bn of operating income against a guided $20bn–$24bn.

Second, the year-ago operating income is understated. Amazon's $17.4bn in Q3 2025 included a $2.5bn charge for a legal settlement with the Federal Trade Commission and $1.8bn of estimated severance costs. Add those back and the underlying figure was about $21.7bn. Against that base, the $22.5bn–$26.5bn guide implies growth of roughly 4% to 22%, not the 29% to 52% a straight comparison suggests.

Why the headline growth rate will look like a slowdown

Amazon's second-quarter sales grew 20%. Its third-quarter guide is 9% to 12%. The gap is mostly the calendar.

Amazon moved Prime Day 2026 to 23–26 June, out of its usual July slot. In 2025 the event fell in July, inside the third quarter. So this year the event's sales land in Q2 and are missing from Q3, while the year-ago Q3 still contains them. Both quarters are distorted, in opposite directions.

Amazon quantified the effect: excluding the Prime Day timing, its third-quarter growth guide would be about 400 basis points higher, or roughly 13% to 16%. That is the figure to compare with the 13% Amazon grew in Q3 2025. A headline growth rate near 10% would therefore say little about consumer demand by itself. What would say something is the North America and International segment growth rates once the shift is accounted for, and what the fourth-quarter guide says about the holiday season.

AWS: the growth rate the market trades

Amazon does not guide AWS separately, which leaves the bar to be set by its own history. AWS sales grew 36.7% in the second quarter to $42.2bn — its fastest growth in 18 quarters, an annualised run rate of $169bn — and AWS operating income rose to $16.6bn from $10.2bn. That acceleration was the main reason the shares rose 15.3% the next day, as we covered in our July results analysis.

The base for the comparison is $33.0bn of AWS sales in Q3 2025, when the unit grew 20%. Some reference points:

Implied sales calculated from Amazon's reported $33.0bn Q3 2025 AWS revenue.

The constraint is supply, not demand, according to Amazon. On the Q2 call Andy Jassy said Amazon would not meet all its demand this year and expected the same in 2027, and Amazon raised its 2026 capital expenditure plan to $220bn from $200bn, citing higher memory chip prices. In that setting, AWS growth is partly a measure of how much new capacity came online in the quarter. A miss blamed on capacity timing would read as revenue delayed, not lost; a miss with softer language on demand would read very differently.

The other figure to watch is AWS operating margin, which was 39.4% in Q2. Servers and memory bought at higher prices are capitalised and reach the income statement gradually as depreciation, so the cost of the $220bn plan arrives with a lag. A margin that slips while growth holds would be that lag showing up; a margin that holds would mean revenue is still outrunning it.

The Anthropic line, and why headline EPS is unreliable

Amazon's Q2 net income was $62.6bn, or $5.75 a share, against $18.2bn a year earlier. Most of the difference was $53.4bn of non-operating pre-tax other income, which Amazon attributed "primarily" to its investments in Anthropic. The year-ago Q3 EPS of $1.95 also contained $9.5bn of pre-tax Anthropic gains.

Those gains are recorded when there is an observable change in the price of Amazon's stake — a funding round, for instance — and nothing is recorded when there isn't one. Analysts cannot forecast that, which is why consensus EPS near $2 should be read as an estimate of the operating business only. A headline EPS far above it may simply mean another revaluation; it would say nothing about AWS or retail. We set out the mechanics, and what Anthropic's IPO filing changes about them, in why Anthropic's paper loss is Amazon's paper gain.

The same caution applies to free cash flow in the other direction. Trailing-twelve-month free cash flow was an outflow of $7.6bn at the end of Q2, against an $18.2bn inflow a year earlier, driven by a $66.1bn increase in purchases of property and equipment. With the capex plan raised, that line is likely to stay under pressure, and how management describes funding — Jassy declined to give details on the Q2 call — is one of the questions the call has to answer.

The three readings, and what each one touches

The third row is the reaction the market has shown most recently. On 22 October the Nasdaq Composite fell 2.6% after beats from Meta and Tesla, with chip stocks down most, as investors focused on spending rather than results — see our analysis of that session. Whether Amazon repeats that pattern depends less on the headline figures than on the relationship between AWS revenue and the capital spent to produce it.

What it touches beyond the stock

NAS100 and US500. Amazon is one of the largest weights in both indices, so its after-hours move goes straight into futures. Because Microsoft and Alphabet will already have reported, the reaction will also reflect how AWS growth compares with Azure and Google Cloud.

The AI supply chain. Amazon's $220bn capex plan is revenue for chipmakers, memory producers, server builders and power suppliers. A change to that plan can reprice those groups more sharply than Amazon itself.

The dollar. The results do not move the US dollar in any dependable way. The link runs the other way: Amazon's guide includes an unfavourable currency effect of about 80 basis points on growth. What matters for the valuation of long-duration growth stocks is the real-rate path, which is set by the Fed a day earlier. Rates are one of the five factors the meter scores across eight currencies — more on how it works on the about page.

Educational macro context only — not investment advice.