BILI: Sell-down scare overblown? Gaming to the rescue

After the HK close on May 19, Bilibili reported Q1 2026 results, delivering another ad-led quarter broadly in line with expectations.

Key takeaways below.

1. Ads remain the clear bright spot: Q1 ad revenue rose 30% YoY, slightly beating estimates despite a weak industry backdrop. Benefiting from strong demand in AI and gaming and a more competitive ad market, Bilibili captured meaningful tailwinds.

This high-growth phase should persist for a while, supported by solid end-market momentum and rising competition in those verticals. It also reflects incremental ad inventory yet to be unlocked on the platform.

2. User growth slowed, but engagement improved: MAUs rose by 10 mn QoQ to 376 mn, roughly back to Q3 last year and short of a new seasonal high, a minor blemish. Holiday timing effects around CNY/winter break may have played a role; management commentary on the call will be key.

User stickiness continued to strengthen YoY, with DAU/MAU at 30.6% and avg. daily time at a record 119 minutes. According to QuestMobile, total user time growth in Jan showed a mild rebound vs. Q4, and given an unchallenging base in last year's Q1, the overall trend looks stable.

Relative to social and long-form peers, Bilibili held up well outside of relentless gains at Douyin and a slight rebound at WeChat. Xiaohongshu has slowed notably, and Kuaishou continues to soften.

2. New game cycle approaching: Q1 was still under a tough comp, with game revenue down 12% YoY. The quarter mainly saw the Hong Kong/Macau/Taiwan release of 'Sanmou' into smaller markets, delivering a steady performance.

As 'Three Kingdoms: Hundred Generals' targets a late-Q2 launch and two additional titles are slated for H2, the new game cycle is drawing closer. Deferred revenue at end-Q1 was up 19% YoY but edged down QoQ, likely due to last quarter's boost from the hit 'Escape from Tarkov'.

3. Live-streaming pressure persists; long-form weakness broadened: VAS growth slowed further to 3.7% YoY in Q1, with both live-streaming and premium membership weaker. Premium members fell by nearly 0.5 mn QoQ.

QuestMobile data show iQIYI, Tencent Video and Youku saw continued declines in total time in Q1. Beyond the drag from AI motion comics, short-form dramas and short video, the holiday box office was also weak, with total Q1 box office down 51% YoY vs. last year's 'Ne Zha' outperformance.

That said, Bilibili's premium content curation differs from those platforms. It relies on a constant supply of UGC from creators and a library of classic global titles, where comments and bullet-screen data help surface user preferences and enable lower-cost catalog buys.

By contrast, iQIYI/Tencent Video/Youku depend on fresh licensed series; when new content is lacking, churn rises.

4. Heavier AI spend is starting to show: OP came in at RMB 170 mn with a 2.2% OPM. While Q1 margins usually dip QoQ on e-commerce off-season ads, marketing seasonality and annual employee payouts, the QoQ decline was steeper than usual.

The difference is increased AI investment flagged last quarter, reflected in R&D reversing three years of tightening and rising 9% YoY. Adj. net income reached RMB 590 mn (mainly adding back SBC at 4% of revenue), for a 7.8% margin, down nearly 300 bps QoQ.

Management previously guided that incremental AI investment, partially offset by other cost controls, would still reduce profit by RMB 0.5–1.0 bn this year.

5. Buyback completed; watch for a new plan: In Q1 the company repurchased 2.5 mn shares for US$60 mn at an avg. price of US$24. By end-Q1, the 2-year US$200 mn authorization approved in 2024 had been fully executed.

Net cash stood at RMB 19.3 bn (US$2.8 bn), leaving room for further buybacks. We will watch management's shareholder return plans on the call, especially after Tencent's remarks about selling higher-valued investment assets to fund its own buyback spooked the market and pressured the stock; Tencent currently holds ~10% of Bilibili.

6. KPIs at a glance

Dolphin Research view

Q1 was again led by ads, but profits are now feeling the drag from AI spend before the new game cycle arrives. Management guided last quarter that this year's incremental AI spend would trim profit by RMB 0.5–1.0 bn. Based on the trend this quarter, we estimate the impact at roughly RMB 50–100 mn, tracking guidance.

Although management also stressed disciplined, phased investment, the market sold first and asked questions later. That is understandable: with Bilibili nearing a US$11 bn market cap at the time, the stock was on ~25x PE on prior profit expectations (adding back SBC), well above peers.

The market had been extrapolating a 15%–20% long-term OPM, as guided by management, into a three-year profit growth trajectory. New AI spend clearly disrupted that path, as it is not a one-off.

Unlike cloud or model-subscription revenue, the revenue lift from AI is less immediately visible, inviting questions on ROI. In the near term, the margin drag extends the timeline to target margins, warranting a higher risk discount.

From today's vantage point, we think the profit drag from AI spend may evolve versus last quarter's view. If shares correct further near term, there could be opportunity:

(1) New game cycle nearing: Revenue growth is the best antidote to concerns about profit erosion from higher spend, and Bilibili still relies on games for top-line acceleration. We are approaching the mid-year launch of 'Three Kingdoms: Hundred Generals', the Q3 launch of 'Shine, Rumi', and a potential year-end launch of 'The Romance of the Three Kingdoms: Wangdao Tianxia'.

While these may have less obvious blockbuster potential than 'Sanmou', H2 comps ease. With incremental titles, the profit drag from AI spend should lessen.

(2) Premium valuation partly reset: On its Q1 call last week, Tencent addressed how it would sustain shareholder returns amid higher capex, noting it may rebalance its investment portfolio to fund buybacks by monetizing certain stakes. Management said it would compare investee valuations with Tencent's own (core biz ~10x PE ex-investments) and consider trimming positions trading at a substantial premium to itself.

As a higher-multiple Chinese ADR/HK name, Bilibili became a perceived target, falling a cumulative 15% Thu–Fri. Tencent's stake fell from 13.5% to 10.5% during 2020–2022, with little activity since 2023.

We think the probability that Bilibili becomes a disposal target for Tencent is low:

First, Bilibili's market cap has mostly traded in the US$8–10 bn range over the past year. Even a neutral near-term fair value just above US$10 bn and an optimistic swing to US$15 bn would imply that selling Tencent's 10% stake would raise under HK$10 bn, barely a month's run-rate at a HK$500 mn/day buyback pace.

Second, as a key channel in the broader entertainment ecosystem—and now a primary channel for AI spend—Bilibili remains strategically relevant. Large-scale disposals would not be ideal for Tencent's partnership strategy.

Thus, we see recent fears as somewhat overdone in the near to medium term. With a closing market cap of US$8.1 bn yesterday, assuming an RMB 0.8 bn profit impact from AI, Bilibili trades on ~21x 2026E adj. PE (adj. net income est. RMB 2.6 bn).

While that is not low on a cross-sectional basis and 2026 is a year of game pressure, the new game cycle beginning in H2 could restore growth and help further digest the multiple.

Detailed analysis follows

I. User growth slowed; engagement strengthened

Bilibili added 10 mn MAUs QoQ in Q1, but did not set a new seasonal high. Engagement kept rising, with DAU/MAU up to 30.6% and avg. time at 119 minutes, +11 minutes YoY.

Per QuestMobile, Bilibili's total time growth in Q1 ranked mid-to-upper among social peers.

II. Ads are again the standout

Q1 ad revenue was RMB 2.6 bn, +30% YoY, modestly above expectations and re-accelerating. Growth was driven by higher total user time (+19% YoY; slightly faster vs. Q4's 18.6%) and improved ad load.

Ad load is still around 7%–8%, leaving room to grow vs. peers. In early Apr, the company added a 'pause ads' slot on the playback page that auto-plays when users pause videos, though it can be disabled by both users and creators.

III. New game cycle nearing

Game revenue remained under a high base and pipeline gap, down 12% YoY in Q1, in line with expectations. The quarter mainly featured the HK/Macau/Taiwan rollout of 'Sanmou' in smaller markets, with a steady showing.

Deferred revenue fell QoQ against the trend, likely due to weaker VAS and softer game momentum.

On the forward pipeline: In addition to last quarter's disclosures, 'Three Kingdoms: Hundred Generals' is planned for late Q2, and 'Shine, Rumi!' for H2. A new co-published title, 'The Romance of the Three Kingdoms: Wangdao Tianxia', is expected in Q4.

'Wangdao Tianxia' was first revealed on Mar 18 with pre-registration across platforms and began its 'Taoyuan' test on Mar 28. It is a 3D sandbox SLG based on the classic 'Romance of the Three Kingdoms' IP that pioneered the SLG genre.

Beyond core strategy gameplay, it emphasizes immersive role-play and progression, leaning more single-player. It also follows a lower-grind, lower-spend approach, simplifying interactions and focusing on strategic battles.

Some institutions project first-12-month revenue of RMB 1.7 bn for this title, about 40% of 'Sanmou'. We see that as somewhat aggressive and note potential cannibalization of the existing 'Sanmou' audience, though the approaching new cycle should bring the inflection point for Bilibili's games closer.

IV. VAS under pressure

VAS revenue, primarily live-streaming and premium memberships, grew 3.7% YoY in Q1, with growth slowing again. Premium subs fell by 0.5 mn QoQ to 24.77 mn.

Outside premium, fandom monetization and live-streaming likely saw the bigger drag from live-streaming.

V. AI spend is showing up

While profit grew strongly YoY, the impact of AI investment began to surface. OP (GP minus opex) was RMB 170 mn with a 2.2% OPM, and adj. net income was RMB 590 mn (mainly adding back SBC at 4% of revenue) for an 8% margin.

1) GPM continued to grind higher

GPM improvement is mainly driven by sustained high growth in the high-margin ad business.

Within COGS, revenue-sharing remains the largest component at 38% of revenue, tied to games, live-streaming and the Huahuo ad platform. Q1 revenue-sharing costs rose to RMB 2.85 bn, up 7% YoY.

2) R&D rose for the first time in three years

On opex, sales was flat and G&A slightly up, with the main increase in R&D. R&D saw a notable +9% YoY rise, the first since 2023, reflecting AI investment.

Although profits were somewhat diluted, monetization efficiency kept improving on our long-watched 'traffic monetization vs. cost' lens, reflecting a stable ecosystem and stronger internal monetization. Whether AI spend is worth it will still require management to deliver faster growth and better monetization.

Past Dolphin Research pieces on 'Bilibili'

Earnings season (recent)

Mar 5, 2026 Call Notes 'Bilibili (Trans): Plans to reinvest part of incremental profit in AI this year'

Mar 5, 2026 Earnings Take 'Bilibili: Ads resilient, a tough 'one-leg' phase for the platform'

Nov 15, 2025 Call Notes 'Bilibili (Trans): Content pay to be a key growth engine'

Nov 15, 2025 Earnings Take 'Bilibili: A second act?'

Aug 21, 2025 Call Notes 'Bilibili (Trans): Reaffirming 15%–20% long-term OPM target'

Aug 21, 2025 Earnings Take 'Bilibili: After 'Sanmou' euphoria, what's next?'

Risk disclosure and statement: Dolphin Research Disclaimer and General Disclosure