META after its Q3 2026 call
Third-quarter sales of $64.6B beat our $63.1B forecast by 2.3%, and adjusted earnings of $7.40 beat our $6.86 by 7.8%. After the release, sales forecasts rose on a higher holiday guide while earnings barely moved, as heavier spending absorbs the gains.
The call, in brief
Takeaways
Revenue grew 26% to $64.60B, with the same growth once currency swings are set aside.
Revenue was $64.60 billion, an increase of 26% year-over-year on both a reported and constant currency basis.
A $9.89B charge to settle teen mental health lawsuits with state attorneys general pushed costs sharply higher.
Third quarter 2026 costs and expenses includes a $9.89 billion charge related to this agreement.
Without that charge, operating income would have been $22.88B instead of $12.99B.
Income from operations of $12.99 billion would have increased by $9.89 billion to $22.88 billion.
Heavy data center spending turned free cash flow negative at $2.12B, despite $32.79B of cash from operations.
Cash flow from operating activities was $32.79 billion, and free cash flow was negative $2.12 billion.
Meta is creating non-voting shares and giving one to every shareholder for each share held, in effect a two-for-one split.
The Board has also declared a one-time stock dividend of one share of Class C common stock for each outstanding share of Class A common stock and Class B common stock
What management stressed
- Muse personal agent takes off
Meta's new personal assistant signed up over 10 million people in the U.S. and Canada in about six weeks.
In just a month and a half, more than 10 million people in the U.S. and Canada have set up Muse personal agents.
- How Muse will make money
Muse stays mostly free. Meta plans to earn fees from businesses on the purchases it helps arrange, plus paid plans.
For most people, the product should be free and we will build commerce tools so we get paid a fee based on the transactions that we help facilitate.
- Cutting Muse running costs
Muse is costly to run, but Meta expects to make it more than ten times cheaper within a year.
we expect to improve its cost efficiency by more than 10x in the next year
- Selling models to businesses
Meta will sell its advanced models to other companies, including a coding tool, to get more out of costly training.
Muse Wonder and even more advanced models that we're training can serve additional high margin services beyond the Muse network, including our API, Muse Code, and other enterprise services.
- Smaller glasses and headset losses
Losses in the glasses and virtual reality unit should shrink to about $15B next year as spending shifts elsewhere.
expect losses to narrow next year to around $15 billion
- Creator ads growing fast
Ads that businesses run with creators nearly doubled and now bring in over $15B a year at the current pace.
Our partnership ads product grew over 97% year-over-year and surpassed $15 billion in annual revenue run rate.
What analysts asked
- What limits Muse's user growth1 question
Muse is expensive to run, so it has not launched worldwide yet. Making it cheaper to run is the key limit.
one of the biggest constraints so far is that because this is a compute intensive service to run, we have not launched it globally yet
- Shopping through personal agents2 questions
Most businesses are eager. Over time, agents will deal with businesses through approved connections rather than browsing websites like a person.
this will move towards more permissioned, structured interactions.
- New models improving ads1 question
The pipeline of ranking and recommendation improvements is healthy, with new research and ad-making tools adding more room to grow.
I think the ranking and recommendations pipeline is healthy.
- Muse at the company's center1 question
Personal superintelligence has been the company's guiding goal for over a year. These are long-term bets on models, agents and glasses.
these are long-term bets.
- Muse cost per user1 question
No figures were given. Spare computing capacity, smaller models and more efficient software should bring costs down.
we don't have sort of specific numbers to share around the cost structure of Muse today.
- How soon Muse makes money1 question
Sooner than usual for Meta, because both the costs and the shopping opportunity are so large.
we are going to work on monetization somewhat sooner on this than our normal playbook
Each line quotes the release or the call. The call's text is our machine transcription of the company's webcast.
What changed after the call
| Number | Before | After | Why |
|---|---|---|---|
| Q4 2026 sales | $72.2B | $75.0B+3.9% | The company guided fourth-quarter sales to $73B to $76B, above what the models had expected.We expect fourth quarter 2026 total revenue to be in the range of $73-76 billion. |
| Q4 2026 adjusted EPS | $8.37 | $8.23−1.7% | The raised full-year expense range leaves heavy fourth-quarter costs, trimming earnings despite higher sales.We now expect full year 2026 total expenses to be in the range of $175-179 billion. |
| Q1 2027 sales | — | $68.6B | New: the quarter after the next report. |
| Q1 2027 adjusted EPS | — | $6.13 | New: the quarter after the next report. |
| FY2026 sales | $252.4B | $256.7B+1.7% | Third-quarter sales beat at $64.6B and the higher holiday guide lifted the year.Q3 total revenue was $64.6 billion, up 26% on both a reported and constant currency basis. |
| FY2026 adjusted EPS | $31.84 | $32.24+1.3% | Third-quarter earnings excluding the settlement charge came in at $7.40, above the forecast.•Diluted EPS of $3.83 would have increased by $3.57 to $7.40. |
| FY2027 sales | $296.0B | $308.8B+4.3% | Management said it can sustain strong revenue growth through 2027, now from a higher 2026 base.Looking ahead, although we are not providing a full-year 2027 revenue outlook, the benefits our AI investments are delivering to the core business, combined with emerging opportunities from model advances and product launches, position us to sustain strong overall revenue growth throughout 2027. |
| FY2027 adjusted EPS | $31.43 | $31.42−0.1% | Capacity is being maximized for 2027, so higher depreciation and cloud costs absorb the extra sales.Consequently, our current plans are geared towards maximizing 2026 and 2027 capacity. |
| FY2028 sales | $339.5B | $361.3B+6.4% | Early Muse adoption and the higher 2027 base lifted 2028 sales views.In just a month and a half, more than 10 million people in the U.S. and Canada have set up Muse personal agents. |
| FY2028 adjusted EPS | $37.03 | $37.05+0.1% | Rising capital spending and the depreciation that follows offset higher sales, leaving earnings flat.We anticipate 2026 capital expenditures, including principal payments on finance leases, to be in the range of $135-145 billion, narrowed from our prior outlook of $130-145 billion. |
Before is our forecast frozen before the release: the average of 6 models as of 2026-10-21 07:41:37 ET. After is the average of the 6 new forecasts. Each why is the editor's line, with the quote that shows it.
FY2027 sales, each model
| Model | Before | After | Change |
|---|---|---|---|
| Opus 5.5 | $290.0B | $306.0B | +5.5% |
| GPT-6.1 Sol | $307.0B | $316.0B | +2.9% |
| Grok 4.7 | $294.0B | $308.0B | +4.8% |
| GLM 5.3 | $292.0B | $308.0B | +5.5% |
| Gemini 4 Argon | $298.0B | $310.0B | +4.0% |
| Muse Wonder | $295.0B | $305.0B | +3.4% |
The outlook, rolled forward one quarter
The outlook page is written again from the new forecasts, with every part of the page before; it is also the next preview.
Our record
| Quarter | Our forecast | Outlook | Reported | Our error |
|---|---|---|---|---|
| Q3 2026 sales | $63.1B | $61.0B–$64.0B | $64.6B | −2.3% |
| Q3 2026 adjusted EPS | $6.86 | — | $7.40 | −7.3% |
The frozen forecast becomes one graded row: our error is our forecast against what was reported. The record grows by one row a quarter.