Dolphin Research

Wielding 'first principles' to disrupt the global aerospace industry

Jun 12 | Dolphin Research Focus: 🐬 Stock #1, $BABA-W(09988.HK) — Reports say Alibaba plans to bid $1.5bn for front-warehouse grocer Pupu. Earlier chatter suggested Meituan and JD were in the running, but both have denied it. Pupu generates approx. RMB 30bn in annual revenue, is steadily profitable, and focuses on 30‑min on-demand delivery across multiple South China cities.

0611 | Dolphin Research Focus: 🐬 Macro/Industry 1) US May CPI +4.2% YoY, up from Apr's 3.8%; core CPI edged up to 2.9% YoY, with energy the main driver. The print was in line with expectations, prompting markets to push back the first cut and trim the number of cuts priced for this year; UST yields moved higher and the USD strengthened.

Below is Dolphin Research's compiled Trans of $Oracle(ORCL.US) FQ4 2026 earnings call. For the earnings analysis, see 'ORCL Plunge? AI Infra Can't Fix the Core Weakness — High Rates, High Debt + Stalled Software'.

Among the 'New Cloud' cohort, $Oracle(ORCL.US) is one of the most watched and debated. On Jun 11 after hours, it reported FY26 Q4 results for the period ended May.

Overall, the print was mixed. OCI growth accelerated, while the broader legacy software portfolio slowed.

GPM improved QoQ off the trough, but still missed estimates by a wide margin. Next-quarter guidance was above expectations.

However, the FY27 full-year outlook was relatively conservative. Net-net, the quarter felt muted...

Oracle 4Q26 First Take: the company reported results for the quarter ended May this morning; overall muted, with few bright spots.

1) On growth, the key metric—OCI revenue—rose 92% in constant currency. It accelerated QoQ as expected, but landed squarely in line with the Street with no upside surprise.

Total revenue was about 19.2bn, up 20.6% YoY. Growth ticked up modestly on OCI strength, or roughly +200bps ex‑FX, and came in slightly above consensus.

2) On new orders, RPO increased by 85bn QoQ to 638bn. With no chatter about fresh megadeals ahead of earnings, the Street was looking for about 590bn.

3) GPM missed again at 65%. It improved QoQ and shows signs of bottoming, but trailed the ~66% Street view.

4) Capex was roughly 16.5bn, down from 18.6bn in the prior quarter. Back‑solving from management’s prior full‑year capex guide of 50bn, the market had penciled in only about 11bn for the quarter, but a sharp pullback would be inconsistent with accelerating OCI demand.

On financing, management said it completed 43bn in bonds and 5bn in equity raises in FY26. For FY27, it plans another 20bn each in debt and equity (already disclosed), effectively securing most of next year’s capex.

5) Guidance: for next quarter, cloud (IaaS+SaaS) growth midpoint is 60%, signaling further acceleration. That is modestly ahead of the ~57% Street.

The rub is the FY27 full‑year outlook: revenue is still guided at 90bn, and the updated full‑year Non‑GAAP EPS of $8.05 is likewise in line with the Street. Both match current market numbers.

This is the final installment in our WMT series. We focus on assessing the stock's investment value.

0610 | Dolphin Research Focus: 🐬 Macro/Industry

This is the third installment of our Walmart deep dive, focusing on Sam's Club — long ignored by investors yet effectively Walmart's third growth engine — and a hard comparison against Costco, the global benchmark for warehouse clubs.

0609 | Dolphin Focus: 🐬 Macro/Industry 1) The US DoD updated its Section 1260H list of Chinese military companies, covering internet, semis, NEVs and biopharma. New additions include Alibaba, BIDU, WuXi AppTec, BYD and Nio, while Huawei, Tencent and Hikvision were already on the list; listed firms issued statements denying defense-related biz and said they will appeal. The list restricts US Gov. procurement rather than imposing financial sanctions, so near-term operating impact is limited but it dampens risk appetite.

In the prior piece, we unpacked Walmart's low-cost engine: township-level, differentiated site selection; an end-to-end, self-run supply chain; reverse pricing with EDLP; and a broad private-label portfolio, showing how 'extreme efficiency + extreme cost discipline' made it the king of offline retail.

0608 | Dolphin Focus: 🐬 Macro/Industry 1) HK and A-shares suffered a broad selloff today. A blowout US NFP flipped rate-cut expectations and pushed UST yields higher, with the tech-led rout on Wall St. spilling into APAC; Korea hit a circuit breaker and Japan tumbled, amplifying fear.

Below is Dolphin Research's Trans from Lululemon's FY26 Q1 earnings call.

Revenue barely in line. Guidance slashed.

LULU 1Q26 First Take: another print with revenue fine but guidance a bomb.

1) Growth improved QoQ but remains soft.

By category, women’s (the core) delivered $1.56bn, +4.4% YoY.

Men’s was $580mn, +6.8% YoY, outpacing women’s.

2) The biggest problem remains a soft North America.

Mainland China was the bright spot: revenue +29%, comps +20%, clearly driven by traffic rather than discounting, making the growth higher quality than in prior quarters.

3) Profit was weak, and guidance was cut sharply.

On opex, earlier-timed brand activations (Indian Wells tennis, Milan Olympics, etc.) and spend tied to distribution-rights competition lifted the SG&A ratio by 310bps to 42.9%.

0604 | Dolphin Research focus: 🐬 Stock pick 1, $Space Exploration Tech(SPCX.US) Elon Musk's SpaceX is set to list on Nasdaq on Jun 12 with ticker 'SPCX'. The IPO is priced at $135, targeting $75bn of proceeds and valuing the company at $1.77tn.

The deal would set a new global IPO record by proceeds. This mega listing could siphon liquidity from global markets, while lifting valuations across the commercial space value chain upstream and downstream.

Below is Dolphin Research's Trans of $Broadcom(AVGO.US) FY26 Q2 earnings call. For our earnings take, please see 'AVGO: Giants vs. Giants — Is the ASIC camp splitting?'

Core highlights from the quarter include: 1) shareholder returns and 2) Q3 outlook. The company paid $3.1bn in cash dividends in Q2, equivalent to a quarterly cash dividend of $0.65 per common share, and guides Q3 consolidated revenue of $29.4bn (+84% YoY)...

AVGO posted FY2026 Q2 results (quarter ended Apr 2026) after-hours on Jun 4 Beijing time.

Management guided AI revenue to $56bn for this fiscal year and over $100bn for next fiscal year, both fairly uninspiring.

AVGO First Take: revenue and GPM were broadly in line with the Street. Growth was mainly driven by AI.Ex acquisition amortization and restructuring, underlying GPM was 76%, roughly flat QoQ. Over the medium term, a higher mix of lower-margin ASICs could still weigh on overall margins.

For next quarter, management guides revenue to $29.4bn, up $7.2bn QoQ. This is slightly above the Street at $28.7bn, with growth largely from AI.

Leverage (Total Debt/LTM Adj. EBITDA) fell to 1.8x. The debt impact from the VMware acquisition has been absorbed within two years.

AI remains the focal point: the company delivered $10.8bn of AI revenue this quarter, up $2.4bn QoQ. It guides next quarter AI revenue to $16.0bn, up $5.2bn QoQ, below the Street’s $17.0bn.

Because the Anthropic partnership is shifting from full-rack purchases to chip sales, the Street had already trimmed Anthropic’s contribution to AVGO. Even so, management’s guide, while showing faster AI growth, is still below mainstream houses’ lowered expectations.

On the other hand, management guides AI revenue of $56bn for this FY and $100bn+ for next FY, which is unexciting. In fact, most Street models are already at $130bn+ for next FY.

Overall, AVGO’s AI biz. is still accelerating. Backed by orders from Google’s TPU program and Anthropic, and by its strength on the networking side, it should continue to gain AI share. The market is bullish on its high growth, but guidance in this print was underwhelming.

Against high-growth expectations, a relatively soft guide may weigh on sentiment and valuation near term. AI revenues are still accelerating, and the story in AI semis and connectivity remains intact. For more, follow Dolphin Research’s upcoming First Take and Trans. $Broadcom(AVGO.US)

Since late 2022, when ChatGPT burst onto the scene, AI has unleashed wave after wave of semiconductor supercycle opportunities. From compute (GPUs) and storage to coordination and scheduling (CPUs), the buildout has already minted a string of $1tn market-cap companies.

0603 ｜Dolphin Research focus: 🐬 Stock 1, $Microsoft(MSFT.US) announced seven in-house, full-stack MAI models at the Jun 2 Microsoft Build developer conference, spanning reasoning, coding, and multimodal vision. The flagship reasoning model, MAI-Thinking-1, is positioned against Claude, while coding performance is on par with Anthropic's top-tier version.

Ride-hailing leader $DiDi(DIDIY.US) reported Q1 2026 results pre-market on Jun 2.

Didi 1Q26 First Take: headline results were weak, with adj. EBITA under RMB 200 mn amid heavy investment.However, vs. expectations the print was better — domestic profit release far exceeded forecasts, and Intl growth was strong while losses came in slightly lower than expected (still sizable in absolute terms).

Breakdown:

1) Domestic GTV rose 10% YoY, with growth easing QoQ but slightly above Bloomberg consensus; order growth showed a similar ~1ppt QoQ deceleration, indicating steady momentum.The standout was profit release, with domestic adj. EBITA near RMB 4 bn, well above the market's ~RMB 3.5 bn.Adj. EBITA margin was 4.6% on GTV, up 60bps YoY.

Domestic platform sales grew ~22% YoY, far outpacing GTV and revenue growth of ~10%.This implies a meaningful uplift in net take rate after driver revenue share and other direct costs, rising from 20.6% to 22.8%.

2) Intl was the highlight on growth, with GTV up 59% YoY; ex-FX, +49%, accelerating by over 10ppt QoQ.Orders rose 27% YoY, with only ~2ppt acceleration and well below GTV growth.Mix shifted toward higher-ticket food delivery, lifting GTV more than orders.

Beyond solid growth, Intl adj. EBITA loss was ~RMB 2.9 bn, still large but better than the market's ~RMB 3.1 bn loss estimate and Q4's RMB 3.4 bn.In other words, loss margin narrowed alongside strong growth.

3) Overall, domestic growth was steady while Intl growth was strong; domestic profit release expanded, and Intl losses narrowed.Company revenue maintained ~10% growth.Absolute profit was still under RMB 200 mn, but vs. the market's expected RMB 500 mn loss, this was a clear beat. $DiDi(DIDIY.US)

Full-stack AI bellwether $Alphabet(GOOGL.US) has seen some recent softness in its share price.

Over the past year, Google raised over $85bn via bond offerings across six markets. This new raise is the largest single round to date.

As for use of proceeds, management outlined a broad plan. The 80bn will mainly go to scaling AI infrastructure, concurrently buying call options to partially hedge future dilution from convertible preferreds, and covering taxes due when employee options vest...

0601 | Dolphin Research Watchlist: 🐬 Macro/Industry 1) China’s NEV market May delivery data is out, with Leapmotor, Nio, and ZEEKR posting sharp MoM gains as production and sales sentiment improves. Li Auto and XPeng saw YoY declines in delivery volumes, while BYD’s monthly sales were steady. Performance among leading OEMs is clearly diverging.