Alibaba is expected to report its June-quarter results (Q1 FY27) before the U.S. open on Thursday, August 20, though the company has not yet confirmed the date publicly on its investor relations site. The stock closed Friday at $125.22, down 16% year to date and sitting roughly 28% below its January high of $173.68. Wall Street has a consensus Strong Buy and an average 12-month target of $190, implying 52% upside from here. The options market prices a 6.6% swing in either direction after results. That implied move is below BABA’s historical post-earnings average of about 7.6% over the last eight quarters. The market is expecting the number, not a shock. What it is not fully expecting is evidence that the open-weight AI monetization model just changed.

Analyst Targets

• Morgan Stanley: Overweight, $180 target (lowered from $190); expects Cloud revenue +45% YoY, margins improving to ~11%
• Barclays: Overweight; sees faster Cloud growth and AI annual recurring revenue above targets, quick-commerce losses narrowing faster than expected
• JPMorgan: Buy, $230 target; based on approximately 12x FY2028 earnings projections
• Jefferies (Thomas Chong): Buy, $230 target; top pick for 2026; cites AI, cloud, and digital consumption growth
• Consensus (40 analysts, S&P Global): Strong Buy, $190.11 average target

Company Profile

Alibaba operates across three principal revenue pillars. China E-Commerce, anchored by Taobao and Tmall, remains the largest segment and the margin engine when it is not being subsidized. The Cloud Intelligence Group, the fastest-growing segment, handles public cloud, AI infrastructure, model-as-a-service, and enterprise AI applications under the Qwen brand. International Digital Commerce, including AliExpress and Lazada, rounds out the portfolio.

Beneath those three sits a strategic bet that now defines the company’s financial profile: Alibaba has pledged at least RMB 380 billion ($53 billion) in cloud and AI infrastructure spending over three fiscal years. Last quarter alone, capex for cloud and AI came in at approximately $5.4 billion. That commitment is why EPS is declining even as cloud revenue accelerates. It is also why the stock has lagged every analyst model.

The Numbers: What Consensus Expects Thursday

• Revenue: RMB 268.5 billion (~$39.8 billion), up approximately 8% YoY (Investing.com consensus: RMB 268.86 billion)
• Adjusted EPS: RMB 10.8 per ADS (~$1.49 USD), down approximately 27% YoY
• Adjusted Net Income: RMB 25.47 billion, down nearly 28% YoY
• Cloud Intelligence Group Revenue: Morgan Stanley projects +45% YoY, ahead of broader market expectations
• Quick Commerce (Taobao Instant Commerce): Expected sequential improvement in unit economics after strong growth in FY26

The EPS decline is not a mystery. It is the explicit, deliberate consequence of a company burning capital to build infrastructure. The question for Thursday is whether management can show the burn rate is peaking while the top line accelerates.

Why Thursday Is Different From the Last Four Quarters

BABA has missed adjusted EPS estimates in four consecutive quarters. Over the past eight quarters, the company has missed adjusted EPS, revenue, and adjusted net income more than half the time. That record matters. It means consensus may still not be conservative enough, and any miss this week arrives with a shorter leash than usual.

But the context heading into this quarter is materially different from what it was six months ago, and the change did not come from inside the earnings report.

On August 3, Alibaba launched Qwen3.8-Max: a 2.4-trillion-parameter Mixture-of-Experts model with 95 billion active parameters at inference time and a 1-million-token context window. The model launched at $2.00 per million input tokens and $6.00 per million output tokens through third-party model gateways, and Alibaba Cloud’s own Model Studio pricing varies by region and deployment. Shares rose about 4.2% in premarket trading in New York and 7% in Hong Kong on the day. Then, within the same week, Reuters reported that Alibaba plans to require large commercial users of the open-weight version of Qwen3.8-Max to share a portion of the revenue they generate from it. Terms have not been finalized.

That second development is the one that matters most going into Thursday’s call.

The Monetization Shift Nobody Has Priced

Until now, Alibaba’s open-weight Qwen releases have often been distributed under permissive terms such as Apache 2.0, though some flagship and specialized variants have used Alibaba’s own Qwen license frameworks with additional conditions. Enterprises running Qwen on their own infrastructure typically owed Alibaba nothing under the permissive releases. Qwen accumulated over a billion cumulative downloads on Hugging Face. The reach was enormous. The direct revenue from that reach was not. Alibaba’s AI monetization story has lived almost entirely on the hosted cloud side: model inference, API calls billed through Model Studio, and enterprise AI contracts counted in the Cloud Intelligence Group’s annualized run rate.

Qwen3.8-Max changes that calculus. If Alibaba does attach commercial conditions to an open-weight release at the large-scale enterprise tier, it would be a shift in posture, regardless of where the final threshold lands. The specific revenue-share rate for large commercial users has not been published, but the structure has been compared in market commentary to Moonshot AI’s approach on Kimi K3, which includes a requirement for larger commercial users to reach a separate agreement.

If Alibaba confirms any version of that structure on Thursday’s call, it reframes the entire Qwen ecosystem. Roughly one billion downloads becomes a potential licensing base, not just a distribution metric. Cloud Intelligence Group AI-related product revenue was running at an annualized rate of approximately RMB 35.8 billion exiting FY26 and accounted for 30% of external cloud revenue in the March quarter. Management also guided that AI-related products could exceed 50% of external cloud revenue in about a year. A revenue-share mechanism layered on top of the existing API and hosted model business could accelerate that trajectory significantly.

Thursday’s earnings call will be the first opportunity for management to put formal parameters around this. That conversation is more important than the EPS number.

Macro and Industry Context

China’s macro environment has not cooperated. The country set its 2026 GDP growth target at 4.5% to 5%, among the lowest in decades, and Chinese consumer spending has remained soft throughout the year. The instant commerce war between Alibaba, Meituan, and JD continues to require heavy subsidies. Core e-commerce profitability in the March quarter dropped 40% year over year as Taobao Instant Commerce investments accelerated.

On the regulatory front, Alibaba settled a $600 million Department of Justice non-prosecution agreement in July over illegal pharmaceutical sales on AliExpress and Alibaba.com between 2016 and 2024. The resolution removes an overhang that had weighed on sentiment, though a separate securities class action lawsuit filed by shareholders alleging failures to disclose a potential Chinese military company classification remains active and has not been resolved.

U.S. AI chip export controls continue to constrain Alibaba’s ability to scale GPU infrastructure using advanced American chips. The company has responded by developing its own AI chips and positioning itself as what CEO Eddie Wu described as the only AI cloud provider in China capable of delivering self-developed AI chips at scale. That claim, if it holds in the June quarter, is not just a marketing line. It is a structural advantage in a market where compute access is increasingly rationed.

Forward Scenarios

Bull Case

Cloud Intelligence Group comes in at or above 45% growth. Management announces firm terms on the Qwen3.8-Max commercial license for large-scale users, including a revenue-share threshold. Quick-commerce unit economics show sequential improvement significant enough for management to set a timeline for the segment reaching breakeven. AI ARR run-rate crosses or approaches RMB 50 billion. The stock re-rates toward $145 to $155 on the session.

Base Case

Cloud grows 38% to 42% YoY, in line with the prior quarter’s trajectory. Revenue hits the low end of RMB 268 billion. Management reaffirms the Qwen monetization direction without formal terms, citing ongoing negotiations. EPS misses modestly but the miss is offset by improved cloud margin commentary. Stock moves 4% to 6% in either direction based on the cloud number and the quality of the Qwen licensing discussion.

Bear Case

EPS misses consensus by a wide margin, continuing the four-quarter streak. Cloud comes in below 35% growth, suggesting the acceleration Morgan Stanley projected has stalled. Quick-commerce losses hold steady or worsen, and management walks back the unit economics timeline. The Qwen revenue-share terms are delayed or vague. The stock tests $110 to $115.

Technical Overlay

BABA closed Friday at $125.22. The 52-week range runs from $91.99 to $173.68. The stock broke above its 50-day and 100-day EMAs in early August on the Qwen3.8-Max launch, reclaiming resistance at $121.41. The current RSI is elevated, reported near 71.7 on recent sessions, which puts the stock technically overbought heading into the earnings release. A beat that confirms the Qwen licensing direction likely pushes the stock toward the $134 to $140 range, where Fibonacci resistance sits. A miss pulls it back toward the $115 to $118 zone, roughly in line with recent moving average support.

What Investors Should Watch Thursday

• Cloud Intelligence Group external revenue growth rate: Whether it accelerates to 45% or stays near 40% is the single most important number in the report.
• AI-related revenue as a percentage of external cloud revenue: Management guided this to cross 50% in about a year. Any update to that timeline matters.
• Qwen licensing terms: Any formal announcement of revenue-share conditions for commercial users of open-weight Qwen models is the most consequential disclosure management could make this quarter. It does not appear in any consensus model.
• Quick-commerce EBITA trend: The pace of improvement in Taobao Instant Commerce unit economics is the main variable for when e-commerce margin recovers.
• Capex guidance: Whether the $53 billion three-year commitment is being front-loaded or spread determines how long the EPS pressure continues.
• Management commentary on the DOJ settlement’s impact on compliance costs and AliExpress revenue.

Bottom Line

The earnings debate around Alibaba has been framed as cloud growth versus margin compression for three quarters. That frame is too narrow going into Thursday. The real inflection is whether Alibaba can close the loop on the Qwen ecosystem: one billion downloads that currently generate no direct revenue, a flagship model now priced commercially, and a revenue-share mechanism that, if confirmed, turns the open-weight strategy from a distribution play into an income stream.

Cloud growing 45% would be a record for the unit and would justify the capex. A Qwen licensing announcement would be a genuine re-rating event. The market is pricing for a modest beat and some reassurance on quick-commerce. If management delivers both the cloud number and the Qwen monetization structure on the same call, the 52% upside implied by the consensus target starts to look less like a long-term aspiration and more like a 12-month problem.

The miss streak, the stock’s RSI, and the ongoing class action are reasons for caution. But the structural setup heading into August 20 is more interesting than it has been at any Alibaba earnings call in the past two years.

For informational purposes only.