Alphabet Q3 2026: Why EPS Fell as Operating Profit Rose

Alphabet Inc. (NASDAQ: GOOGL, GOOG) reported diluted earnings per share of $2.12 for Q3 2026, down from $2.87 a year earlier. Yet operating income increased 30%, to $40.576 billion. The apparent contradiction is largely explained by investment gains and losses below the operating line — a distinction that matters when investors assess whether Google's core businesses actually deteriorated.¹

The income statement tells two different stories

Alphabet's September-quarter revenue rose from $102.346 billion to $127.577 billion. Operating profit rose by $9.348 billion. But other income (expense), net swung from a $12.759 billion gain in Q3 2025 to an $8.221 billion expense in Q3 2026. That $20.980 billion pre-tax swing more than offset the additional operating profit.¹²

Why did Alphabet's EPS fall despite higher operating profit?

Alphabet's reported net loss on equity securities in other income was $8.664 billion in Q3 2026, versus a $10.734 billion gain in Q3 2025. That is a $19.398 billion year-on-year change in this single line. Alphabet's earnings release explicitly estimates that the Q3 2026 equity-securities loss reduced diluted EPS by $0.55 after the associated tax effect.¹²

The SEC note is more precise than the popular shorthand of calling the entire loss a paper loss. It reports $12.436 billion in unrealized losses on marketable and other equity securities, partially offset by $3.027 billion in unrealized gains on non-marketable holdings and $0.745 billion in realized gains on equity-security sales. The net equity-security result thus includes both realized and unrealized items. The filing attributes the other-income decline primarily to marketable equity movements largely from SpaceX and smaller unrealized gains on certain non-marketable holdings.²

Did Alphabet's core operating business weaken?

Based on the operating line, not across the board. Clarifo's SEC-XBRL financial history records $127.577 billion of revenue and $40.576 billion of operating income for the quarter ended September 30, 2026. Google's Services segment earned $39.157 billion of operating income, while Google Cloud earned $11.121 billion, compared with $3.594 billion a year earlier. Some costs sit in separate Alphabet-level activities; the segment profits should not be added without that reconciliation.¹²

Those figures make a blanket claim that Google's operating engine collapsed difficult to defend. But the counterargument — that the quarter was excellent and the market should ignore the EPS miss — also skips real risks. Search & other revenue growth slowed to 12%, the monetization of AI Mode remains less mature, and infrastructure investment absorbs cash before revenue and margins can fully prove its economics.¹³

Cash flow is the harder test than an EPS adjustment

Alphabet's Q3 2026 operating cash flow was $42.290 billion and property-and-equipment purchases were $51.355 billion. The resulting company-defined free cash flow was negative $9.065 billion. An unrealized investment mark can depress accounting profit without immediately consuming the same amount of operating cash; expanding data centers, however, do consume cash when paid for. Investors should not confuse the two effects.¹

Even here, a single quarter is insufficient. The company's trailing-twelve-month free cash flow through Q3 2026 was still positive $19.747 billion, despite negative free cash flow in Q2 and Q3. Management expects significantly higher investment in technical infrastructure, and said full-year 2026 capital expenditures were expected to fall in a $195 billion–$205 billion range. These are different concepts: realized historical free cash flow versus management's future investment guidance.¹³

Does the reported EPS decline make Alphabet stock cheaper?

A valuation case built only on trailing GAAP EPS can be distorted by unusually large equity marks. But removing those marks without also evaluating interest costs, taxes, future dilution and capital needs would create another distortion. A defensible analysis starts with separate operating-profit, non-operating-investment and cash-flow forecasts. It then states the price, date and expected return assumptions explicitly. This article does not label the shares cheap or expensive without a complete, time-aligned valuation.

Financing choices matter as well: in Q3 2026, Alphabet issued approximately $12.4 billion of Class A and Class C shares through its at-the-market program and $28.7 billion of senior unsecured notes, while reporting no common-share repurchases. Those are reported financing decisions rather than forecasts of AI returns. They also make per-share dilution and future financing costs relevant to the equity valuation.¹

There is a second relevant filing detail: the Q3 10-Q reports significant future data-center and energy commitments, including $130.5 billion of future payments on leases that had not yet commenced as of September 30, 2026. Those future lease payments are not the same as current-period capex and should not be simply added to the quarter's cash outflow. They do show why assessing long-duration commitments matters when modelling the AI investment cycle.²

For context on the broader investment cycle, see our comparison of AI infrastructure spending and cash generation across Alphabet and other hyperscalers. It helps distinguish the cash costs of building capacity from the current accounting effects of holding equity investments.

We would revisit this analysis after the next interim report by tracking operating income, the bridge through other income, total share count, quarterly free cash flow, financing costs and the capex outlook. The evidence supports a narrower conclusion than either Reddit camp: reported EPS fell materially, but operating profit grew. That helps explain the quarter; it does not resolve the valuation argument.