Comcast reported third-quarter 2026 results on Thursday, October 22, before the open. Revenue fell 6.1% - and on a pro forma basis rose 0.5%.
And that is the whole report in one line: to know what happened here you have to hold three comparison bases at once, and two of them contradict the third.
There may be errors or inaccuracies. Found something that looks wrong? Write to me and I will fix it.
What was reported
| (millions of dollars, except per share) | Q3 2026 | Q3 2025 | Change |
|---|---|---|---|
| Revenue | 29,298 | 31,198 | -6.1% |
| Pro forma revenue | 29,298 | 29,151 | +0.5% |
| Net income attributable to Comcast | 3,448 | 3,332 | +3.5% |
| Adjusted net income | 3,824 | 4,125 | -7.3% |
| Adjusted EBITDA | 9,106 | 9,669 | -5.8% |
| Pro forma adjusted EBITDA | 9,106 | 9,038 | +0.8% |
| Earnings per share (dollars) | 0.97 | 0.90 | +7.2% |
| Adjusted earnings per share (dollars) | 1.07 | 1.12 | -4.0% |
| Net cash provided by operating activities | 7,902 | 8,693 | -9.1% |
| Free cash flow | 4,211 | 4,945 | -14.8% |
Source: Comcast's third-quarter 2026 press release, October 22, 2026, filed with the SEC on Form 8-K as Exhibit 99.1. Adjusted EBITDA, adjusted net income, adjusted earnings per share, free cash flow and the pro forma basis are non-GAAP measures, as the company defines them.
‼ Why two bases, and what the gap between them says.
The pro forma basis strips out two completed disposals: the Versant separation, completed January 2, 2026, and the sale of Sky's operations in Germany, completed May 31, 2026.
So the 6.1% is not a business that contracted by 6.1% - it is largely businesses that are no longer here. Excluding them, revenue and adjusted EBITDA were, in the company's words, "consistent with the prior year period".
And that leads to a conclusion that suits nobody: Comcast did not fall 6%, but it did not grow either. 0.5% in revenue and 0.8% in EBITDA is standing still.
‼ And here is what the pro forma basis does not hide
This is, in my view, the important line in the report, and it is not affected by disposals at all - the Connectivity & Platforms segment:
| (millions of dollars) | Q3 2026 | Q3 2025 | Change |
|---|---|---|---|
| Residential Connectivity & Platforms | 17,018 | 17,636 | -3.5% |
| Business Services Connectivity | 2,645 | 2,576 | +2.7% |
| Total segment revenue | 19,663 | 20,212 | -2.7% |
| Adjusted EBITDA - Residential | 6,110 | 6,484 | -5.8% |
| Adjusted EBITDA - Business Services | 1,478 | 1,454 | +1.7% |
| Total segment adjusted EBITDA | 7,588 | 7,938 | -4.4% |
| EBITDA margin - Residential | 35.9% | 36.8% | -90 bps |
| EBITDA margin - Business Services | 55.9% | 56.4% | -50 bps |
| EBITDA margin - total segment | 38.6% | 39.3% | -70 bps |
Residential connectivity is 86.5% of segment revenue, and its revenue fell 3.5% and its EBITDA 5.8%. That is not a disposal and not a currency effect - that is the core.
And margins eroded across all three lines. The company describes this as "the investment in our go-to-market strategy" - that is, discounts and incentives to hold customers. That is a legitimate description of a business choice, but the numerical result is falling revenue at falling profitability.
Wireless, which is growing
| Q3 2026 | |
|---|---|
| Net line additions in the quarter | 368,000 |
| Line additions over the last twelve months | 1.6 million |
| Total wireless lines | 10.6 million |
| Penetration of domestic residential broadband customers | about 18% |
| Penetration of total addressable lines in the footprint | 7% |
And the figure that makes this interesting is actually the penetration figure: 10.6 million lines are only about 18% of Comcast's own domestic residential broadband customers. Which means 82% of the customers already paying it for internet do not buy wireless from it.
The company also reports that nearly 45% of its broadband customers are already on Gig-Plus speed tiers - up nearly 10 percentage points year over year - and that premium unlimited plans represented approximately 40% of postpaid phone connects. It also launched Xfinity Shield, a cybersecurity and home protection platform layered onto its Wi-Fi.
Peacock and the studios - the side that worked
Peacock posted adjusted EBITDA of $261 million on revenue of $1.8 billion - a second consecutive profitable quarter. The year-over-year improvement is $478 million, which means the business lost $217 million in the year-ago quarter. Paid subscribers rose 17% to 48 million.
And the studios: The Odyssey, released in July, became Universal's highest-grossing film of all time with nearly $1.8 billion at the box office year to date, and Universal was the first studio to pass $5 billion at the global box office in 2026. Studios EBITDA rose 18.6% to $402 million.
Against that, the company notes near-term cyclical headwinds in Theme Parks, without giving that a separate figure in the release. It opened Fast & Furious: Hollywood Drift at Universal Studios Hollywood.
Cash flow and the separation
Free cash flow fell 14.8% to $4,211 million, and cash from operating activities 9.1% to $7,902 million. Capital expenditures were consistent at $3.1 billion: in Connectivity & Platforms they rose 3.6% to $2.4 billion, primarily on customer premise equipment, and in Content & Experiences they fell 10.6% to $639 million. The company paid $1.2 billion in dividends.
Co-chief executives Brian L. Roberts and Mike Cavanagh write that the separation of NBCUniversal and Sky is progressing and on track, and that it should create two independent companies. The release contains no target date.
The stock
The report was published before the open, and the investor call was held at 8:30 a.m. Eastern. So the trading reaction is that same day, October 22.
The stock closed at $22.84, up 3.44% from $22.08. It opened at $22.44, rose as high as $23.29 - 5.48% above the prior close - and fell as low as $22.00, meaning it also touched below that close. Volume: 54.4 million shares, 35% above the day before.
On Friday the stock fell 0.66%, to $22.69.
הזווית שלי
דעה אישית של אילן אברמוב - לא ייעוץ ולא המלצה
There are three stories in this report, and they do not point the same way - and in my view that is itself the story.
The first story is the pro forma basis, and it is fair. A company that sold Sky in Germany and separated Versant is not "down 6%" - that is a meaningless comparison. The adjustment is correct and legitimate. But its result is 0.5% in revenue and 0.8% in EBITDA, and that says the remaining business is standing still. Pro forma here is not a defensive instrument; it is an instrument that replaces a fall with stagnation.
The second story is the one I cannot dress up: the residential connectivity core. 86.5% of Connectivity & Platforms revenue, revenue down 3.5%, EBITDA down 5.8%, and margin down 90 basis points. When revenue falls and profitability falls faster, that is not the price of growth - it is the price of holding on. "Investment in our go-to-market strategy" is the company's phrasing for paying in order not to lose customers.
And the third story is that Peacock and the studios genuinely worked. A $478 million swing in Peacock's profitability in one year is a real change, and 48 million paid subscribers is a base. A film that passed $1.8 billion at the box office is an event, and I note explicitly that it is an event - a studio cannot plan an annual blockbuster, and $402 million of EBITDA resting on one film is not a run rate.
And what connects the three, in my view: the growing part of Comcast is wireless, and there it has penetrated only 18% of the customers it already has. That is both the clear opportunity and the admission - 82% of its broadband customers choose a different wireless provider.
What I will follow: residential connectivity revenue and EBITDA, which are the core and are not hidden by any comparison basis; and wireless penetration, the one measure here with large and obvious room.
(To be clear: this is my personal opinion only. Nothing here is a recommendation to take any action, and there is no forecast here of the price of any security.)
Ilan Abramov, the author, or anyone on his behalf may hold some of the securities or funds mentioned in this article.
