Spider-Man Turns Sony's Beat Operational, but the Stock Got There First

Results vs. Consensus

Sony reports in yen under IFRS, and the number the market trades is group operating income. Two consensus bases are relevant and they are not interchangeable: the yen operating-income poll, which Japanese and global institutional investors position against, and the ADR earnings-per-share line, which the U.S. listing screens on. Both are shown below. Where no reliable published consensus exists for a line, the cell reads "n/a" rather than a manufactured number.

Q2 Scorecard

The yen consensus for the quarter had already risen by roughly ¥40bn since August, when the Street carried the September quarter at about ¥465bn. Spider-Man: Brand New Day opened on July 31, and the estimate climbed with its box office. Sony still cleared the higher bar by ¥117.3bn. Sales were reported above estimates without a published figure, so the cell carries no magnitude. The quarterly operating cash flow is derived by subtracting the first quarter's ¥197.4bn from the six-month statement.

Year-Over-Year Comparison

Every income-statement figure above is taken from the Consolidated Financial Summary filed on Form 6-K, not from figures quoted on the call. Prior-year comparatives are the continuing-operations basis that excludes the Financial Services business spun off on October 1, 2025; last year's quarter also carried ¥50.5bn of net income from discontinued operations that is excluded here. Comprehensive income fell because the yen strengthened from ¥162.2 to ¥157.1 to the dollar between the June and September quarter-ends, producing a ¥185.8bn translation loss on foreign operations. That is a balance-sheet effect, not an operating one.

Sequential Comparison

The sequential step-up is the better guide to what changed inside the year. The June quarter's ¥476.5bn contained roughly ¥56bn of tariff refund; the September quarter's ¥622.3bn contains an undisclosed residual of it, at most roughly ¥24bn, and it absorbed a ¥30bn Sony Honda Mobility loss on top. A ¥145.8bn sequential gain built on a smaller refund and a larger charge is the first evidence in three quarters that Sony's operating line is rising on its own content rather than on items outside the businesses.

Revenue. Sales of ¥3,429.4bn grew 10.3%, but Sony's segment currency disclosures sum to ¥212.6bn against a ¥321.5bn increase, which leaves growth of roughly 3.5% excluding currency. Pictures contributed ¥167.7bn of the increase on its own. By customer location, the United States grew 20.2% and Europe 14.5%, the two regions where the film and the game slate sold; China fell 6.0% and Japan was flat, which is consistent with the smartphone and camera weakness Sony describes in its hardware segments.

Margins. Operating margin reached 18.1% against a clean 15.7% a year ago, and gross margin rose from 32.4% to 37.0%. The mix did most of the work: Pictures went from a 4.0% to an 18.0% margin, and gaming from a clean 15.3% to 18.2%. Selling, general and administrative expense rose 13.8%, faster than sales, which is what a quarter with a global theatrical tentpole and a major first-party launch should look like. Music was the one large segment whose margin fell, from 21.3% to 20.5%, for a second consecutive quarter. Inside operating income, the equity-method loss widened from ¥1.6bn to ¥22.8bn, which is where most of the Sony Honda Mobility charge sits.

Earnings and cash. Diluted earnings per share rose 47.9% against a 44.4% increase in attributable net income; the gap is a 2.4% lower diluted share count from the buyback. The effective tax rate eased from 27.9% to 26.7%. The bigger change is cash. Operating cash flow for the quarter was roughly ¥457.9bn, nearly double last year's, and six-month operating cash flow of ¥655.4bn is up 33.1% even though income taxes paid rose from ¥89.9bn to ¥252.0bn. The conversion problem we flagged in August looks, on this evidence, more like tax timing than like a deterioration in the businesses.

Segment Performance

Sony reports five segments plus a corporate line. Segment sales are stated before intersegment elimination and segment operating income excludes unallocated corporate expense, so the segment columns do not foot to the consolidated line; the consolidated row below is the income-statement figure, not the sum of the segments.

Q2 Segment Detail

Adjusted OIBDA by Segment

Sony's Adjusted OIBDA adds back depreciation and amortisation other than film costs, broadcasting rights and internally developed game content, then removes items the company deems non-recurring. It neutralises last year's gaming charges, which makes it the cleaner year-on-year read for that segment.

Revenue by Product Category (external customers)

Pictures

The segment that made the quarter. Sales rose 48.5% to ¥513.7bn, or 37% on the U.S. dollar basis Sony uses for this business, and operating income rose roughly sevenfold to ¥92.6bn, a second-quarter record. On the dollar basis, operating income was $581M against $93M. Theatrical revenue of $1,276M against $114M is the whole story: Spider-Man: Brand New Day, released July 31, had taken $952M in North America and $1,548M internationally by September 30.

"Spider-Man: Brand New Day, released in July, was a record-breaking hit with worldwide cumulative box office revenue exceeding 2.5 billion U.S. dollars. It is the highest-grossing film of all time in North America and the third highest-grossing film of all time globally." — Lin Tao, Executive Officer and CFO

Underneath the film, the segment is weaker than the headline suggests. Television Productions revenue fell 31% in dollars to $483M, Media Networks slipped 2% to $896M, and the motion-picture streaming-licence line fell to $186M from $237M. Sony raised the full-year Pictures operating income guide by a third, to ¥200bn, but that still implies a second half of ¥82.6bn against a clean ¥99.5bn a year ago, with Jumanji: Open World in December carrying launch marketing.

Assessment: This is what the bull case on Sony's content portfolio looks like when it works, and it worked unusually well. It is also the hardest profit in the group to underwrite forward. The FY2028 slate has two large titles, The Legend of Zelda on April 30 and Spider-Man: Beyond the Spider-Verse on June 18, plus post-theatrical windows for this film, so the segment need not collapse. But the incremental ¥78.7bn is a hit, not a run-rate, and we model it that way.

Game & Network Services

Sales rose 8.6% to ¥1,208.5bn, of which ¥73.0bn is currency, so sales excluding currency grew about 2%. Operating income rose 82.8% to ¥220.1bn, a second-quarter record, and 29% against last year's clean base. The composition is the encouraging part. Digital software revenue rose 23.3% and network services 19.5%, while hardware revenue fell 4.9% on 2.6 million PlayStation 5 units against 4.0 million. First-party software units rose from 6.3 million to 8.2 million on Marvel's Wolverine, which sold through 2.4 million copies in its first 50 days.

"With regard to PlayStation (PS5) hardware, we have been able to offset the rise in memory prices through adjustments to promotions and pricing, as well as additional cost-reduction, and we expect hardware profitability for the current fiscal year to remain on par with the previous fiscal year." — Lin Tao, Executive Officer and CFO

Engagement turned. Monthly active users reached 127 million in September, up 7% and a September record, against +2% in the June quarter, and total playtime rose 4% after falling 4% in the June quarter. Sony raised the segment's full-year operating income guide ¥40bn to ¥700bn, citing third-party software, network services and lower costs, partly offset by moving God of War Laufey from February into the first quarter of FY2028.

Assessment: The best-quality profit in the release. It comes from software and subscriptions on roughly 98 million PlayStation 5 consoles sold in to date, not from refunds. Two caveats keep us from extrapolating it. Hardware sell-in has now fallen 35% or more for three consecutive quarters, and part of the guidance raise is next-generation development cost that is moving into next year rather than going away (Key Topic 3).

Music

Sales rose 10.3% to ¥598.4bn, of which ¥34.3bn is currency, and operating income rose 6.4% to ¥122.8bn, a second-quarter record despite the absence of last year's Demon Slayer theatrical release, which shows up as a 14.6% decline in Visual Media & Platform revenue. On the dollar basis that removes translation, streaming revenue grew 7% in Recorded Music and 6% in Music Publishing, down from 10% and 8% in the June quarter and from the 9% to 14% band we cited at initiation.

"We believe such evergreen catalogs are assets that will continue to generate value over the long term, even in the age of AI." — Lin Tao, Executive Officer and CFO

The prepared remarks leaned on catalog: songs older than 18 months made up more than half of streams on Spotify's Global Top 200 through September, and Michael Jackson's catalog went from roughly 100 million to more than 3 billion streams on that chart after the biopic. The balance sheet tells the investment side. The Recognition Music Group transaction, closed July 15, brought in ¥542.7bn of music catalogs and ¥308.7bn of long-term debt secured by them, for ¥256.4bn of cash consideration and ¥66.3bn of third-party non-controlling interest. Music's long-lived assets rose 22% in the quarter to ¥2,513.7bn.

Assessment: Still the highest-quality earnings stream in the group, and still the one whose returns are drifting the wrong way. Margin fell for a second quarter, dollar-basis streaming growth slowed again, and the segment return-on-invested-capital table in the supplement has not been updated since May, even though the denominator just grew by more than a fifth. The guide went up ¥25bn to ¥445bn; the return on the capital behind it is the number Sony is not showing.

Imaging & Sensing Solutions

A record quarterly operating income of ¥154.6bn on falling sales. Revenue fell 6.0% to ¥577.6bn, and 12.7% excluding ¥40.9bn of currency; image sensor revenue fell 7.3%. Operating income rose 11.8%, but currency contributed ¥25.9bn, so it fell about 7% excluding currency. Margin of 26.8% is the highest in the group, helped by mix and by the yen.

"In the smartphone market this quarter, growth declined approximately 20% year-on-year mainly due to the impact of conditions in the memory market. Although the market for mid-range and lower phones declined significantly, we believe the decline in the high-end market was limited." — Lin Tao, Executive Officer and CFO

The physical data confirm the softening. Image sensor wafer input fell to 147 thousand 300mm-equivalent slices a month from 155 thousand a year ago, against capacity of 160 thousand, so utilisation dropped from about 99% to about 92%; Sony attributes part of the dip to the July Kumamoto earthquake. More telling, Sony now forecasts December-quarter capacity of 155 thousand, the lowest in the period we track. The full-year operating income guide rose ¥30bn to ¥450bn on cost reductions and currency, net of an approximately ¥16bn earthquake loss now included.

Assessment: The margin is excellent and the trajectory is not. Sales excluding currency fell by double digits, the implied second-half operating income of ¥173.2bn is 14% below last year's clean ¥201.2bn, capacity is being trimmed, and management, while declining to comment on its largest customer's procurement, did not contest that the customer will diversify supply from next year (Analyst Q&A). The segment is earning record profit at the top of a cycle that its own forecast says is turning.

Entertainment, Technology & Services

Sales slipped 1.4% to ¥567.8bn, about 6% lower excluding currency, and operating income rose 5.1% to ¥64.1bn only because of currency and the U.S. tariff refund, which Sony names as the main driver. Excluding the ¥6.1bn of disclosed currency, operating income fell about 5% before removing the refund. Imaging revenue fell 1.5% on a slowdown in China, Displays rose 1.2% on the new True RGB BRAVIA range, and Sound fell 1.0%. The full-year guide was left at ¥150bn of operating income, which implies a second half 21% below last year's.

One balance-sheet line deserves attention. Segment inventory reached ¥441.9bn, up 20% in the quarter and 18% year-on-year, ahead of a holiday season and the April start of the television joint venture with TCL. Management did not discuss it.

Assessment: The weakest segment underneath, and the one carrying the group's remaining memory cost. The tariff refund is masking a decline that will be visible once the refund laps. The inventory build is either memory bought ahead or product waiting for demand, and the next two quarters will tell which.

All Other and Sony Honda Mobility

All Other lost ¥21.9bn against ¥2.9bn a year ago, and the equity-method line moved from a ¥1.6bn to a ¥22.8bn loss. Sony disclosed an additional loss of approximately ¥30bn related to Sony Honda Mobility, following the ¥44.9bn recorded in the March quarter, and said the recognition of losses from downsizing that venture is substantially complete.

Assessment: If the losses are indeed complete, this is the last of the portfolio charges we have tracked since initiation. Sony's non-recurring schedule does not include this charge, so its "nil" for the quarter understates what was cleaned up (Key Topic 11).

Key KPIs

Key Topics & Management Commentary

Overall Management Tone: Management was confident and, for the first time this year, had no composition problem to explain away; the prepared remarks spent their time on franchises rather than on refunds, charges or currency. Where the call thinned was the future: next year's memory price, the next console and the largest sensor customer's sourcing were each answered with a framework or a deferral rather than a number, and the capital-expenditure increase was not discussed at all. Compared with the June-quarter call, which was about the composition of a beat, this one was about content, and it was noticeably less forthcoming on the year after this one.

1. A Content Quarter: Two Franchises Carried It

Of the ¥135.6bn increase in clean operating income, ¥78.7bn came from Pictures and ¥49.9bn from gaming. Within those, one film and one first-party game did most of the work, alongside a strong third-party software month. Management framed the result as proof of the group's platform-and-content model, and the cross-segment data it cited support that framing: unit sales of related game titles roughly doubled after the film's release, and the Sony Music releases featured in it passed one billion streams.

"We believe that this strong performance was achieved through the symbiotic combination of stable growth in our distribution platforms, including PlayStation Network and Crunchyroll, and hit content driven by our strong IP and creative talent, such as Spider-Man, Wolverine, and Ella Langley." — Lin Tao, Executive Officer and CFO

The same summary slide carried a sentence that is easy to skip. Sony said some of the first-half profit "came from such things as tariff refunds and demand being brought forward," which is a candid description of a half in which the record ¥1,098.8bn of operating income contains roughly ¥56bn to ¥80bn of refunds and an image-sensor shipment pattern that management expects to reverse.

Assessment: The quarter is better than the June quarter in the way that matters to us: it beat on what the businesses did. But it is narrow. Outside Pictures, and before currency, clean operating income grew about 3%, which keeps our third bear point alive even as the headline argues against it.

2. This Raise Is Operational, and the Guide Is Conservative Again

Sony raised full-year operating income ¥150bn to ¥1,870bn, sales ¥500bn to ¥13,000bn, net income ¥120bn to ¥1,330bn and operating cash flow ¥100bn to ¥1,600bn. In July, management said the raise came mainly from tariff refunds and currency. This time it said the effect of tariff refunds and memory on the full-year forecast was largely unchanged.

"The impact of U.S. tariff refunds and memory market conditions on the full-year operating income forecast remains largely unchanged." — Lin Tao, Executive Officer and CFO

The segment drivers confirm it. Gaming rose ¥40bn on third-party software, network services and lower costs. Music rose ¥25bn on streaming revenue. Pictures rose ¥50bn on theatrical revenue. Imaging & Sensing Solutions rose ¥30bn on cost reductions and currency, net of the earthquake. Electronics was unchanged. Currency assumptions for the second half were left at approximately ¥153 to the dollar and ¥175 to the euro, against a first-half average of ¥159.3 and a November 5 rate of about ¥157.7.

Assessment: The raise is real this time, and the guide is still a floor. It implies ¥771.2bn of operating income in the second half against a clean ¥744.9bn a year ago, or 3.5% growth after a first half that grew 32.9% on the same basis. If the yen stays near current levels, Sony's own sensitivities add roughly ¥20bn to the second half before any operating upside.

3. Part of the Gaming Raise Is a Deferral Into Next Year

The prepared remarks attributed part of the ¥40bn gaming raise to "improvement in costs." The investor session established what that means. By management's account it is largely timing rather than efficiency: next-generation development cost that will now be recognised in FY2028 rather than FY2027 because of accrual timing.

"That is because for the next generation title development, timing is different. The next generation investment itself, we are on schedule, but some of them, the accrual basis is different, so that it's going to be on next year's." — Lin Tao, Executive Officer and CFO

In the media session, management added that the next-generation investment is weighted to the second half, that development is on schedule, and that some spending may fall into next year under the accounting rules. Separately, God of War Laufey moves from February to the April-to-June quarter, which shifts a first-party launch out of FY2027 as well.

Assessment: A deferral flatters FY2027 and loads FY2028, the year that already loses the ¥80bn tariff refund. We add the deferred development cost back into our FY2028 cost base. This is the kind of disclosure that only surfaces under questioning, and it changes how much of the gaming raise we would carry forward.

4. TSMC Signed: Sony Controls the Venture, So the Plant Spending Stays on Sony's Books

The commitment we tracked for two quarters was delivered on August 11. Sony Semiconductor Solutions and TSMC signed a definitive agreement to form Advanced Vision Semiconductor Manufacturing Corporation in Koshi City, Kumamoto. Sony will contribute approximately ¥465bn, in cash and by transferring its newly built Koshi fab; TSMC will contribute approximately ¥282bn in cash, in phases. Volume production is expected in 2029. Sony will be the sole controlling shareholder and the venture will be a consolidated subsidiary.

"The TSMC collaboration that we have announced this time, amidst this fierce competition, in order for us to win, we have to work with them and work on both our technical and cost front and win in the market ultimately." — Naoya Horii, Senior Vice President, Corporate Planning and Control

Two structural points follow from the terms. First, Sony holds roughly 62% of the planned contributions and TSMC roughly 38%, so TSMC funds more than a third of the venture's equity and brings its process technology. Second, because the venture is consolidated, its capital expenditure will be reported inside Sony's consolidated capex, with TSMC's money appearing as non-controlling interest. Sony said the agreement's effect on FY2027 results is expected to be immaterial, and the ¥10bn of annual preparation cost added to the segment guide in July was not revisited.

Assessment: The agreement meets the test we set: it has an ownership split, an investment figure and a timeline. It also changes how the fab-light claim should be measured. Reported group capex will not fall because of this venture; what falls is Sony's share of the capital burden. We will judge the pillar on capex net of partner contributions from here, and we would like Sony to report it that way.

5. Group Capex Rises, and Content Spending Comes Back

The supplemental schedule revised FY2027 capital expenditure for the second time in two quarters. Total additions rose from ¥760bn to ¥780bn. Plant and equipment rose ¥5bn to ¥325bn, content-asset additions rose ¥30bn to ¥345bn, reversing two-thirds of July's ¥45bn cut, and right-of-use additions fell ¥15bn to ¥110bn. Imaging & Sensing Solutions additions rose ¥5bn to ¥240bn, with the image sensor component at ¥220bn against ¥215bn in July. That is still 10.8% below FY2026's ¥246.7bn.

"Through collaboration with TSMC in the joint venture, we expect both companies to enhance the performance and cost competitiveness of image sensors and drive the creation of new value in the era of physical AI. We also expect to lay the foundation for further improvement in the capital efficiency of this segment over the long term." — Lin Tao, Executive Officer and CFO

The prepared remarks spoke of capital efficiency "over the long term"; none of the revisions was mentioned, and none was asked about. Since May, plant and equipment guidance has risen from ¥275bn to ¥325bn, an increase of 18%.

Assessment: This is the third of our three upgrade conditions, and it went the wrong way. Sensor capex remains lower than last year, which is the segment-level evidence we wanted. But group plant spending is up 18% since May, total capex is higher than the budget, and Sony has offered no bridge. With the TSMC venture consolidated, the group capex line is now the place where the fab-light claim will be tested, and it is rising.

6. Sensors: Record Profit on Shrinking Volume, and a Customer Expected to Diversify

Imaging & Sensing Solutions earned a record ¥154.6bn on sales that fell 12.7% excluding currency. Management described a smartphone market down about 20% because memory costs are squeezing mid-range and low-end handsets, with the high end holding up better. Mobile sensor shipments tracked the market, and first-half and second-half shipments will split differently from prior years because of channel inventory and customer product-cycle timing.

"I cannot give you the details, but in general, the customers will source from different sources. I think this is normal business practice that we must assume. In line with that, of course, our company have to think about our production capacity in line with the scenarios that we are assuming." — Lin Tao, Executive Officer and CFO

That answer, given to a direct question about the largest customer's share from next year, sits alongside a December-quarter capacity forecast cut to 155 thousand slices a month. The Kumamoto Technology Center is back to pre-earthquake output, and the approximately ¥16bn earthquake loss is now in the guide.

Assessment: The most consequential new fact on the call for FY2028. Sony did not dispute the premise that its largest customer will dual-source, it described planning capacity around that scenario, and it trimmed near-term capacity. Record margins in a segment that is losing volume and preparing for a loss of share at its biggest account are a peak signal, not a base.

7. Memory: Next Year's Volume Is Agreed, Next Year's Price Is Not

Last quarter, every memory answer was scoped to the current year. This quarter Sony moved one step further out, on quantity only.

"Regarding memory procurement for next fiscal year, based on discussions held so far with various suppliers, we have aligned on the quantity of memory needed for our projected hardware unit sales." — Lin Tao, Executive Officer and CFO

Asked whether next year's memory price could be absorbed through pricing, and what a new console would do to gross margin, management said memory prices "will be one of the challenges the management will face" and that it would announce its approach "at appropriate timing in the future." Hardware profitability for FY2027 is guided on par with FY2026, protected by price increases, promotion adjustments and cost reduction.

Assessment: Securing next year's volume removes the supply risk that put this bear point at materialising six months ago, and we move it to contained. The price risk remains open and it is larger now, because it lands in the same year as the next-generation development cost and the end of the tariff refund.

8. PlayStation Engagement Turns Up as the Box Count Keeps Falling

Engagement is the strongest operating signal in the release. Monthly active users rose 7% to 127 million and playtime rose 4%, both improving sharply from the June quarter. Network services revenue grew 19.5%. Management framed the hardware decline as life-cycle normal and the platform as growing.

"The year-on-year trend for both metrics improved significantly compared to the first quarter of this fiscal year, which we believe demonstrates that the baseline of engagement is steadily improving." — Lin Tao, Executive Officer and CFO

The holiday quarter carries Grand Theft Auto VI on November 19, which Sony monetises through digital store revenue recognised gross. Management said it expects the title to contribute for several years, as earlier entries in the series did. Sony also plans a Crunchyroll hub on the PlayStation 5 home screen next spring.

Assessment: Rising engagement on a base that is adding consoles at about two-thirds of last year's pace is a good outcome, and it lengthens the runway before the next platform. The open question remains duration: add-on content grew only 4.4% this quarter, and it is the line that eventually follows installed-base growth.

9. Cash Flow Catches Up With Profit

Operating cash flow was the weak line in the June quarter, down 22% while operating income rose 40%. In the September quarter it nearly doubled to about ¥457.9bn, and the full-year guide rose ¥100bn to ¥1,600bn. Income taxes paid in the half rose by ¥162.1bn, which explains much of the gap between profit growth and cash growth in the half.

"You are right when you say our operating cash flow has been upwardly revised. In the current midterm, our operating cash flow generation capacity has increased steadily." — Lin Tao, Executive Officer and CFO

The guide still implies a sharp second-half slowdown: ¥944.6bn of operating cash flow against ¥1,474.0bn in the same half of FY2026, a 36% decline.

Assessment: A large part of our cash-conversion concern was tax timing, and the commitment we set in August has been partly answered. We are not closing it, because the full-year guide still implies a steep second-half decline that has not been explained.

10. The Balance Sheet Moves to Net Debt as Buybacks Accelerate

Net cash, measured as at initiation (cash against short-term borrowings, current and long-term debt and both lease-liability lines), went from ¥539.2bn in March and roughly ¥299bn in June to net debt of ¥77.5bn at September 30. Long-term debt rose ¥411.6bn in the half, of which ¥308.7bn is debt secured by the acquired music catalogs. Payments for businesses were ¥314.1bn and treasury-stock purchases ¥263.4bn.

"Currently, while we invest in the growth investment, while keeping the financial health, we still have room to increase shareholder returns." — Lin Tao, Executive Officer and CFO

Buyback execution accelerated. By September 30 Sony had repurchased 74.5 million shares for ¥263.4bn, 52.7% of the ¥500bn facility, and by October 31, 86.1 million shares for ¥309.0bn, or 61.8%. Management said strategic investment will continue to centre on entertainment.

Assessment: Shareholder returns are running ahead of the pace we assumed and management is signalling more. The trade-off is now explicit: the fortress balance sheet of the initiation is a modestly levered one, and the next acquisition, Tamron or otherwise, is likely to be funded at least partly with debt.

11. Sony Honda Mobility: A Last Charge, Kept Off the Non-Recurring Schedule

Sony's non-recurring schedule reads nil for a second consecutive quarter, but the quarter was not free of charges.

"In the All Other segment, we recorded an additional loss of approximately 30.0 billion yen related to Sony Honda Mobility during the quarter. This substantially completed the recognition of losses associated with the downsizing of that company's operations." — Lin Tao, Executive Officer and CFO

Together with the March quarter's ¥44.9bn, Sony has now recognised about ¥75bn of losses on the venture in two quarters. The charge was not a surprise: Sony built ¥30bn of further losses into its May forecast. What is new is that it landed in this quarter, inside a record result, and that management now calls the process substantially complete.

Assessment: Closing out the venture is good portfolio housekeeping and supports our simplification pillar. The classification is the problem we raised with the tariff refund in reverse: a one-time recovery was kept on the operating line in June, and a one-time loss was kept on it in September. Before any tariff residual, our clean basis puts the quarter about ¥30bn above Sony's own presentation.

12. Cybersecurity Reaches the PlayStation Account Base

The media session raised a series of security incidents. One questioner said there seemed to be some impact on family accounts on the PlayStation Network, and another cited customer concern on social media about the PlayStation service. Management apologised, declined to give specifics, and said appropriate measures had been taken for affected PlayStation customers.

"First of all, we are aware that there have been a series of incidents, and we as management take this seriously, and we apologize to the customers and stakeholders to whom we have inconvenienced." — Lin Tao, Executive Officer and CFO

Assessment: Not yet a financial issue and not raised by investors, but it concerns the asset this quarter's results were built on: an account base that grew to 127 million and stores payment details for network-services and store revenue. The 2011 network outage is the precedent for what a platform-level breach costs. We will track it as a risk to engagement and to regulatory cost.

Guidance & Outlook

Segment Guidance

Capital Expenditure Guidance

Implied second-half ramp: The ¥1,870bn guide less the ¥1,098.8bn delivered leaves ¥771.2bn for October to March. The comparable half of FY2026 earned ¥678.5bn as reported and ¥744.9bn excluding its net non-recurring items, so the guide implies 3.5% clean growth after a first half that grew 32.9%. By segment, only Music is guided up in the second half; gaming, Pictures, electronics and sensors are each guided below last year's clean second half, and the corporate line is guided to improve sharply as last year's March-quarter charges do not repeat.

Street at: The September-quarter estimate rose from about ¥465bn in August to ¥505bn at the print, which suggests the Street was already positioned above the July guide before it was raised. We did not find a reliable post-print full-year figure. The sell-side commentary we found was positive, but we could not verify its dates against our cutoff.

Guidance style: Conservative, and consistently so for a third consecutive forecast. May set ¥1,600bn, then July raised it to ¥1,720bn and November to ¥1,870bn, each raise after a quarter that beat. The second-half guide holds currency about ¥4.7 below spot against the dollar, includes the full earthquake loss, and assumes every hardware segment declines. We treat ¥1,870bn as the low end of the range. Note, though, that some of the conservatism is deliberate cost timing that moves into FY2028.

Analyst Q&A Highlights

Sony runs two consecutive question sessions after the prepared remarks, one for media and one for investors and analysts, each about twenty minutes. The entries below are drawn from the investor session, where five questioners were taken and each was held to two questions. The call is conducted in Japanese; the exchanges below come from the interpreted English record.

Where the Upside Sits After a Record Half

The first questioner also asked which businesses still have room to grow now that gaming, music and pictures have all been raised, and what will drive the next mid-range plan. Management answered with the ecosystem rather than a segment, and named the TSMC venture as the one change to Sony's growth model.

Q: "The entertainment business, you have upwardly revised, that is gaming, music, and pictures. So, your capability to generate revenue and profit is increasing... So, which businesses do you think will have still room for growth, and what will be the driver?"
— Munakata, Goldman Sachs

A: "First of all, finally, in my opinion, we have been investing on entertainment for many years, and finally, we are seeing mega hits every year, and we will be able to deploy the IP in a very multilateral, multifaceted manner. So, we are now going into that stage."
— Lin Tao, Executive Officer and CFO

Assessment: "Mega hits every year" is the bull thesis in management's own words, and it is a claim about repeatability that this quarter cannot prove. The answer committed to nothing measurable ahead of the next plan, which is where we will look for targets.

Whether Engagement Can Keep Rising After the Holiday Launch

The same questioner asked whether the engagement gains extend beyond this year and which users drive them. Management named monthly active users, playtime and revenue per user as the KPIs, and argued that the largest third-party launch of the year will contribute for several years rather than one quarter.

Q: "Thinking about next year onwards, do you think there will be further upside to the engagement next year onward? What type of users do you think will be more active?"
— Munakata, Goldman Sachs

A: "GTA VI, there will be contribution this fiscal year, but even looking at series games titles in the past, these games have driven the growth for multiple years. So, I think this new title will be contributing even after next year, but this is not the only driver."
— Lin Tao, Executive Officer and CFO

Assessment: A credible answer, because the precedent supports it. Management also named revenue per user alongside engagement as a key KPI. That emphasis is consistent with a platform that is monetising a slower-growing base harder, which is the right strategy for this stage of the cycle.

Next Year's Memory Bill and the New Console

A direct question on whether next year's memory prices can be passed through, and whether a new console would break gross margin, drew the call's clearest non-answer. Management acknowledged the problem and moved to platform health and investment discipline without giving a number.

Q: "The memory prices are rising, and you said that you have secured the necessary volume, but next fiscal year, if you are to introduce a new console, I think that the game profitability will decline as a result of having to make investments. So, the next fiscal year's memory price and forecast, well, is it to the level that you can absorb it from pricing policies?"
— Hirakawa, BofA Securities

A: "As you say, the memory prices will be one of the challenges the management will face... We have to support this soundness of our platform and try to think how rationally we can invest in our platform, and we can announce that at appropriate timing in the future."
— Lin Tao, Executive Officer and CFO

Assessment: The volume is secured and the cost is not, and management chose not to size the cost. The answer is consistent with a FY2028 gaming plan that will carry memory inflation and next-generation spending at the same time, which we now model.

Image Sensor Share at the Largest Customer From Next Year

The same analyst asked how Sony is planning for a largest-customer sensor share below 100% from next year. Sony declined to comment on the customer, but did not contest the premise and described capacity planning around it. The follow-up added that speed and the TSMC partnership are how Sony intends to compete.

Q: "The second, from next fiscal year, image sensor strategic customer share will not be 100%. So, you are communicating with the strategic customers. What sort of scenario can you anticipate?"
— Hirakawa, BofA Securities

A: "I cannot give you the details, but in general, the customers will source from different sources. I think this is normal business practice that we must assume... I think there will be fierce competition, and so what we can control is our technical superiority and cost superiority."
— Lin Tao, Executive Officer and CFO

Assessment: The most important exchange of the session for FY2028. "Fierce competition" at the top of the smartphone sensor market is a new phrase from this management, and it arrived together with a capacity cut. We lower our FY2028 sensor assumptions accordingly.

What the Gaming Cost Improvement Actually Is

A question about the cost improvement behind the gaming raise, and whether it was development, promotion or something else, produced the disclosure that part of it is a timing shift. The same exchange established that hardware units are running above Sony's own plan for the year, without a figure.

Q: "The second question is about for the same gaming industry, about the profit going up, but the cost improvement you have said, but the cost improvement, what kind of a cost improvement? Is it due to the development or is it for the sales promotion or other cost factors?"
— Ayata, JP Morgan Securities

A: "That is because for the next generation title development, timing is different. The next generation investment itself, we are on schedule, but some of them, the accrual basis is different, so that it's going to be on next year's. Because of the lag in the timing of the quarters."
— Lin Tao, Executive Officer and CFO

Assessment: The answer converts part of a guidance raise into a deferral. It is honest, and it was given only because the question was specific. Investors adding the ¥40bn gaming raise to their FY2028 run-rate are adding a cost that is coming back.

Sensor Seasonality and the Wafer-Input Dip

A question on why wafer input fell and how the seasonal pattern of sensor shipments is changing drew a qualitative answer: customer launch timing has moved, this year is a transition, and the full-year volume is not expected to change much. The input decline was attributed to the earthquake.

Q: "Looking at the appendix, the capacity and wafer input, the actual wafer input went down. I think that was impacted by the Kumamoto earthquake. I want to confirm that. Also, how do you perceive seasonality?"
— Katsura, SMBC Nikko Securities

A: "But this fiscal year is a transition period, so that really changed the balance between the first and second half... The input of the wafers have actually gone down, and as you say, this was impacted by the Kumamoto earthquake."
— Naoya Horii, Senior Vice President, Corporate Planning and Control

Assessment: Plausible for the September quarter, but it does not explain the December-quarter capacity forecast of 155 thousand slices, which is below the pre-earthquake level. Read together with the customer answer above, the capacity reduction looks like planning, not repair.

What the Higher Cash Flow Will Fund

The final substantive exchange asked how the higher operating cash flow will be deployed and where strategic investment is heading. Management pointed to growth investment within financial discipline, more shareholder returns, and entertainment as the focus of acquisitions.

Q: "You have upwardly revised by JPY 100 billion. Looking at the business environment of entertainment as a whole, how you will be investing, and how you see the environment, and how you see the potential targets of investments."
— Katsura, SMBC Nikko Securities

A: "Currently, while we invest in the growth investment, while keeping the financial health, we still have room to increase shareholder returns... we will continue to strategically invest mainly in the entertainment business."
— Lin Tao, Executive Officer and CFO

Assessment: The clearest signal yet that a further return programme is coming, likely with the next plan. It also confirms that the acquisition appetite has not been satisfied by the music catalog deal, and with the balance sheet now at net debt, the next deal is likely to need debt.

What They're NOT Saying

1. The group capital expenditure bridge, for a second quarter. Total FY2027 capex guidance rose from ¥760bn to ¥780bn, plant and equipment is now ¥50bn above the May budget, and content-asset additions were raised ¥30bn after being cut ¥45bn in July. None of it was explained, and with the TSMC venture consolidated, this is now the line that decides whether Sony is getting less capital-intensive.
2. How much of the venture's plant spending Sony's reported capex will carry, and when. The agreement sets contributions of about ¥465bn and ¥282bn and production in 2029, but not the spending profile, the timing of TSMC's cash or the size of the government support being sought.
3. The size of the second-quarter tariff refund. Sony says the full-year tariff effect is "largely unchanged" and names refunds as the main driver of the electronics profit increase, but it did not give the September-quarter amount. The cleanest read of underlying growth depends on it.
4. How much next-generation development cost moved into FY2028. Management confirmed that part of the gaming cost improvement is accrual timing, not savings, and gave no figure.
5. What next year's memory will cost. Volume for FY2028 is agreed. Price, and what it does to console margin in a launch year, was deferred to "appropriate timing in the future."
6. The share Sony expects to keep at its largest sensor customer. Management did not contest that the customer will dual-source from next year, described scenario planning, and trimmed December-quarter capacity, but gave no volume or revenue range.
7. Music's return on invested capital after the catalog deal. The segment ROIC table in the supplement still shows the May forecast. Music's long-lived assets rose 22% in the quarter, and the return on them is the number that would show whether the catalog strategy is earning its cost.
8. Crunchyroll subscribers. The supplement repeats the March 31, 2026 figure of over 21 million for a second quarter, in a release that promotes a new Crunchyroll hub on the PlayStation 5 home screen.
9. Bungie's remaining goodwill, for a third consecutive quarter. Gaming goodwill is ¥481.5bn. After ¥120.1bn of impairments in FY2026, the residual exposure has not been addressed since.
10. Tamron. The acquisition proposal disclosed in July was not mentioned in either session, and its terms remain undisclosed.
11. The scope of the security incidents. Management apologised and said measures had been taken, but gave no count of affected PlayStation accounts and no cost estimate.

Market Reaction

• Pre-print setup: The ADR closed at $25.95 on November 4, up 1.4% year-to-date against an S&P 500 up 12.3%, down 7.6% over the trailing twelve months, and up 8.8% over the prior 30 days from $23.85 on October 6. The 52-week closing range entering the print was $19.32 to $30.26, placing the stock 34.3% above its low and 14.2% below its high, and 11.6% above the $23.26 close after our August downgrade.
• Tokyo before the release: Sony released after the Tokyo close on Thursday, November 5. In that day's session, before the results, the ordinary shares rose 7.1% to ¥4,211 while the Nikkei 225 fell 1.8%.
• ADR reaction session (November 5): The ADR gapped up 6.2% to $27.57 at the open, traded a $27.34 to $28.30 range, and closed at $27.54, up 6.1% or $1.59. Volume was 11.5 million shares against a 30-day average of 3.9 million, 2.9 times normal.
• Index and peer context (November 5): The S&P 500 fell 4.4% and the Nasdaq Composite 5.5%, in an AI-led sell-off. Taiwan Semiconductor's ADR fell 10.8%, Microsoft 3.6% and Nintendo's ADR 1.3% after its own results missed; Take-Two rose 2.1% two weeks ahead of the Grand Theft Auto VI launch.
• Tokyo reaction session (November 6): The shares opened 4.5% higher at ¥4,400 and reached ¥4,450, then closed at ¥4,142, down 1.6%, as the Nikkei 225 fell 6.0%. Volume of 47.1 million shares was two-thirds higher than the day before.

The ADR's 6.1% gain is three moves stacked together. About 2.4 points came from Tokyo's pre-release rally, which the ADR had not yet traded. About 3.1 points was the response to the results themselves, measured as the ADR's yen-equivalent close of roughly ¥4,342 against Tokyo's pre-release ¥4,211. About half a point was the yen strengthening against the dollar. The results-driven part, a little over 3%, is a solid but not exceptional response to a 23% operating-income beat and a 9% guidance raise.

The relative performance is the striking part. A 6.1% gain on a day the S&P 500 lost 4.4% is an excess return of about 10.5 points, and Tokyo's 1.6% decline the next day was 4.3 points better than a 6.0% index fall. Some of this is rotation. Sony has been treated as a stock exposed to AI disruption of content, and in the heaviest market sell-off since April 2025 it traded as a non-AI holding with hit-driven earnings and a buyback. That is a positioning effect as much as a verdict on the quarter.

The Tokyo session is the more cautious signal. The shares gave back the entire 4.5% opening gain as the index fell 6.0%, and closed at the equivalent of about $26.27 per ADR, 4.6% below the New York close. That is the level at which this report is written, and it is the anchor that matters for the rating: the stock is now 32% above where we initiated in May and has moved, on the New York close, past the multiple we called full in August.

Street Perspective

Debate: Is Spider-Man a quarter or a franchise?

Bull view: The film proves that Sony's content machine now works across the group, not inside one segment: game sales doubled, the soundtrack passed a billion streams, and the post-theatrical windows will pay for years. With a Zelda film and a Spider-Verse sequel in the next fiscal year, Pictures has a slate, not a single hit.

Bear view: One film produced 58% of the quarter's clean profit increase, and the rest of the group grew about 3% excluding disclosed currency. Television Productions fell 31% in dollars. A record achieved by a top- three film of all time sets a comparison that no slate can be expected to repeat.

Our take: The bear is right about the numbers and the bull is right about the mechanism. We credit the cross-group effect qualitatively and model the incremental ¥78.7bn as a hit. FY2028 Pictures profit can hold near this year's level only if both 2027 tentpoles land.

Debate: Does the TSMC agreement make Sony fab-light?

Bull view: TSMC is putting in about ¥282bn of cash for a minority stake, Sony keeps control and the technology lead, and image sensor capex is still running about 11% below last year. Sony gets leading- edge process technology and a partner funding more than a third of the equity.

Bear view: The venture is consolidated, so its plant spending will sit in Sony's reported capex, and group plant spending is already up 18% since May. The largest sensor customer is expected to dual-source, so the new capacity is being funded into a market where Sony's share is about to fall.

Our take: Both are describing the same deal. Sony's share of the capital burden falls; the group's reported capital intensity does not. We keep the pillar on track on the first measure and want the second reported clearly, because the customer question now matters more than the fab question.

Debate: Can engagement carry PlayStation to the next console?

Bull view: Monthly active users and playtime are rising, network services grew almost 20%, the biggest third-party launch in years arrives on November 19, and FY2028 memory volume is agreed. Sony is monetising a large installed base at higher prices with flat hardware profitability.

Bear view: Console sell-in has fallen by a third or more for three quarters, add-on content is growing in low single digits, and FY2028 brings memory inflation, next-generation development cost that was deferred from this year, and the launch-year economics of a new console.

Our take: The bull has FY2027 and the bear has FY2028. Engagement is the best operating signal in the release, and it is why we raise our gaming estimate. The deferral and memory price are why we do not carry the full raise into next year.

Debate: Is Sony an AI loser or an AI refuge?

Bull view: Content libraries, franchises and a closed gaming platform become more valuable as AI makes generic content abundant, and the market voted that way on November 5, when Sony rose while AI infrastructure stocks collapsed.

Bear view: The same AI boom is starving phone makers of memory, which shrank the smartphone market by a fifth this quarter and is now Sony's sensor problem. A relief rally on a rotation day is not a re-rating on fundamentals.

Our take: Sony is both, in different segments. The memory channel is the one with numbers attached, and it points down for sensors in FY2028. We would not underwrite a higher multiple on a single rotation session.

Model Update & Valuation Framework

In August we carried FY2027 operating income of ¥1,800bn against a company guide of ¥1,720bn. The company now guides to ¥1,870bn, so our above-guide stance was right and has been overtaken again. The table resets our assumptions against the November guide and adds an explicit FY2028 view, because a twelve-month horizon from here runs mostly through that year.

Valuation

At the $27.54 reaction close and the November 5 rate of ¥157.67 to the dollar, the ADR is equivalent to approximately ¥4,342 per ordinary share. On 5,839,845,972 shares outstanding excluding treasury stock at September 30, that is a market capitalisation of approximately ¥25.36tn. Adding the ¥77.5bn of net debt gives an enterprise value of approximately ¥25.44tn.

The August column uses the enterprise value at that date (with roughly ¥299bn of net cash) and puts it on today's earnings basis wherever the row says so, so the multiple rows compare like with like. On the same November guide, the stock has moved from 11.3 times to 13.6 times enterprise value to operating income since we downgraded. On our own FY2027 estimate the move is 11.0 times to 13.2 times. The guide rose 8.7% over the same period; the price rose 18.4%. Most of the move is multiple, and the stock re-rated before the print as much as after it: on the guide in force each day, the pre-print multiple of 14.0 times was higher than the post-print 13.6 times.

Scenario frame, twelve months

We continue to value the group on enterprise value to operating income, because reported net income is still distorted by the spin-off comparatives, the equity-method line and items Sony does not classify as non-recurring. Our FY2028 estimate is close to FY2027, so the multiple is similar on either year.

Implied ADR values use the ¥157.67 rate, the September 30 share count and net debt, and treat the remaining buyback as value-neutral at market prices. They are sensitive to the yen in dollar terms: a move to ¥145 lifts each by roughly 9% with no change to the business, and a move to ¥170 reduces each by roughly 7%.

Valuation impact: We raise our twelve-month target to $28 from $25, on 13.5 times enterprise value to our ¥1,925bn FY2027 operating income estimate. Against our August framework of 12.5 times ¥1,800bn, the enterprise value is 15.5% higher: about seven points from the estimate and eight from a one-turn higher multiple, which we think the completed portfolio clean-up, the signed TSMC terms and the operating cash flow recovery justify. Net debt, the share count and currency bring the per-share increase to 12%. The target is about 2% above the reaction close and about 7% above the level implied by Tokyo's November 6 close; with the 0.8% dividend yield, expected total return is roughly 2.5% to 7.5%.

Thesis Scorecard Post-Earnings

The pillars below are the ones established at initiation and carried in August. They are graded against what this print and call showed, with the status tag we carried into the quarter and where it moves.

Overall: The thesis is stronger on operations and unchanged on capital. This is the first quarter since initiation in which the beat came from the businesses rather than from refunds, currency or the absence of charges, and the memory and cash-flow bear points both improved. Against that, the growth is concentrated in one film and one game, the sensor business is losing volume and preparing for a loss of share, and the capital intensity claim is now harder to measure, not easier. Of the three conditions we set in August for a return to Outperform, two have been met: a signed TSMC agreement with terms, and a quarter that beat on something other than refunds and currency. The third, a group capex bridge, was not provided, and capex was raised.

Action: Hold. We said in August that all three conditions would move us back to Outperform; two arrived and one did not, and the stock rose 18% from our downgrade while they did. We net the two explicitly: the operating evidence raises our estimate 7% and our multiple by one turn, and the price has already moved further than both. Existing holders should stay; we would not add at these levels. For the next quarter we set the netting in advance. We would upgrade on a group capex bridge that shows capital intensity falling on Sony's share, together with either an ADR below roughly $24.50 or a FY2028 plan that supports operating income above ¥1,950bn excluding the refund. We would move to Underperform if the ADR trades above roughly $31, our bull-case value, without an increase in our estimates, or if FY2028 sensor guidance shows the customer change cutting segment profit by more than 10%.