CMCSA · Q4 2025 Earnings

Q2'25 deliberate pricing reset → Q3'25 multi-quarter EBITDA drag, wireless reframed as growth engine → Q4'25 2026 as "largest broadband investment year in history," tone shifts defensive-to-offensive on narrative but the timeline keeps moving out.

The strategic framing has flipped from cost discipline to investment offense in a single quarter, even as the underlying timeline lengthens. Q3 management was apologetic about a "deliberate investment phase" that "will take time and carry a cost." This quarter Brian Roberts opened with: "We're at an inflection point, both in our industry and at Comcast. The business is changing rapidly, competition has never been more intense, and the choices we're making right now matter." The language is more assertive but acknowledges higher stakes — and the EBITDA lapping point has moved from "next several quarters" (Q3) to "second half of 2026" (Q4), which is a deferral, not an acceleration. Management is selling conviction, but the math is requesting more patience.

The broadband revenue line went negative this quarter (-1.1% YoY), but the mechanism matters: ARPU still grew +1.1%, with Jason describing it as "slight growth, but consistent with the deceleration that we had previewed." The pressure on the top line is subscriber attrition (-181K) more than pricing — and that nuance reframes the watch from "is ARPU collapsing?" to "is the new pricing/packaging stabilizing the base?" Combined with the reaffirmed no-rate-increase posture for early 2026, broadband pricing power is intentionally capped until H2 2026 at earliest, with management betting that volume stabilization plus wireless attach carries the segment until the investment cycle laps.

The free-line conversion story has hardened from a soft "intention" (Q3) to an explicit operating dependency (Q4). This quarter: "we expect to convert the vast majority of free lines into paying relationships, which in turn should provide a meaningful tailwind to convergence revenue growth." The promotion in language is also a promotion in risk — H2 2026 conversion is now the load-bearing column under the equity story, and wireless net adds slowed 12% sequentially this quarter as competition stepped up, which is the first sign the free-line top-of-funnel is harder to fill than the prior trajectory implied.

Peacock has shifted from a narrowing-losses narrative to an explicit Q1 2026 dilution warning, and the Q4 loss itself widened materially to $552M (vs. $372M prior year) on NBA rights absorption. Management telegraphed the Q1 NBA absorption peak directly: "the first quarter will be the peak volume period with roughly 50% of our games played, which will also result in peak EBITDA dilution." Pairing the peak-Q1 warning with the FY 2026 "meaningfully improve" framing implies a sharp back-half recovery that the company has not yet quantified — a setup that depends entirely on advertising scaling and affiliate renewals landing on schedule.

The capital allocation tone has moved from "disciplined" (Q3 buyback trim) to "constrained" (Q4). Jason's commentary that "our leverage ratios will increase slightly on the back of the spinoff... our intention will be to migrate back to the 2025 pending leverage of 2.3 times" combined with the disclosure that ~$2B of 2025 cash tax benefits will not recur signals that 2026 free cash flow will be materially lower than 2025's $19.24B. The buyback pace established in Q3 is now bracketed by a deleveraging commitment on one side and a tax-benefit cliff on the other.