Web Research
Figures converted from CNY at historical FX rates — see data/company.json.fx_rates. Ratios, margins, and multiples are unitless and unchanged.
Web Research
The Bottom Line from the Web
The web research layer for this run was unavailable. Every Parallel search phase (industry, warren, quant, sherlock, historian, forensic) and every specialist follow-up query returned HTTP 402 — Insufficient credit before a single page was fetched. Zero queries executed, zero pages were read, zero dossier was produced. The page below therefore does what it can: it inventories the questions the specialists had no way to answer from the filings, marks each as needs external evidence, and flags where this gap leaves the thesis exposed. The single most important consequence is that the largest shareholder — Glory Achievement Fund / "Yi'nan Li" Bull Trust — remains unidentified and the seven 13D/A filings in twelve months remain unexplained.
Web research provider unavailable. Parallel.ai returned Error 402 — Insufficient credit in account across all six preload phases and all nine specialist-query files. No external sources were consumed. Every "finding" below is an explicit open question, not a verified claim.
Queries Executed
Pages Read
Preload Queries Planned
Specialist Queries Planned
What Matters Most
These are the ten questions that would change an investor's read on NIU most, ranked by importance. Each is a question the specialists raised after exhausting the filings; each remains unanswered because the web layer never executed. They are listed in order of how much the answer would move the thesis.
1. Who actually controls 38.4% of the economics and 29.7% of the vote?
Status: unverified — needs external evidence. The "Glory Achievement Fund Limited" Bull Trust appeared on the cap table in December 2023 with a 60.24M Class A position. Its disclosed beneficiary is named "Yi'nan Li," whose romanization is identical to CEO Yan Li's family name. The 20-F states only that "three protectors unrelated to Mr. Li" exercise voting and disposal authority until August 2028. Whether Yi'nan Li is the CEO under a different romanization, a family member, or an unrelated third party determines whether the company has effectively a 35%-economic founder block or a 30%-vote third-party overhang. Sherlock flagged this as the single highest-priority open question in the people analysis.
2. What do the seven 13D/A amendments in twelve months mean?
Status: unverified — needs external evidence. Glory Achievement Fund has filed Schedule 13D/A amendments in July 2025, twice in February 2026 (Feb 2 and Feb 11), and twice in March 2026 (Mar 4 and Mar 13). Five amendments in ten weeks of 2026 is dense filing activity even for a trust under active administration. Whether the amendments reflect (a) share disposals creating overhang, (b) pledges as collateral, (c) protector changes, or (d) administrative-only updates is unknowable from the filings themselves. A single confirmed sale would invalidate the "founder-aligned" reading of the cap table; an administrative update would close the question entirely.
3. Did dealers pre-buy ahead of GB17761-2024, and is Q3 2025 a clean run-rate?
Status: unverified — needs external evidence. Q3 FY2025 was NIU's only profitable quarter in a nine-quarter window: revenue +65.4% YoY and net income $11.5M. The implementation date for the revised GB17761 e-bicycle safety standard was September 2025 — exactly the quarter NIU printed the inflection. If industry-wide dealer pre-buying occurred (as Forensic suspects), Q3 2025 should be excluded from the FY2026 run-rate. If it did not, the operating-leverage thesis has its single best proof point. Forensic ranked this as a medium-priority query because the answer changes how investors annualize the recovery.
4. Is the FY2026 revenue / margin consensus framed?
Status: unverified — needs external evidence. The internal estimates JSON returned an HTTP 429 error. Without sell-side consensus, the page cannot frame whether management's Q2 2026 guide of +25% to +45% revenue is ahead of or behind expectations. The same gap applies to FY26 gross-margin consensus (relevant to whether the 19.6% FY25 print is the new floor or a one-quarter peak) and the FY26 operating-margin path. Quant ranked this as a high-priority open question.
5. How fast are Honda EM1 e:, ACTIVA e:, and QC1 scaling?
Status: unverified — needs external evidence. Honda is the only competitor with simultaneous global brand recognition, dealer footprint, and balance-sheet capacity to attack NIU's premium positioning. Honda EV motorcycle unit volumes are not disclosed at line-item level in the consolidated 20-F. A breakout to 200,000+ units in India, Vietnam, Indonesia, or any commuter-Asia market would be a moat-breaking event for NIU and would shorten the 24–36 month durability runway estimate to under 12. This is the largest external competitive threat the filings cannot quantify.
6. Is Yadea launching a connected-vehicle platform?
Status: unverified — needs external evidence. Yadea Group Holdings (1585.HK) is the China volume leader at ~13M units. The single most diagnostic input to the moat-durability call is whether Yadea is actively copying NIU's smart/IoT stack. A confirmed connected-vehicle launch with national-scale rollout would tighten NIU's premium-ASP compression timeline from 24 months to 12. Yadea also launched a sodium-ion battery scooter in January 2025; the implied BOM cost trajectory matters for NIU's 20–22% gross-margin aspiration.
7. What is Niu App ARPU on subscriptions?
Status: unverified — needs external evidence. "Accessories, spare parts, and services" is 9.4% of FY2025 revenue ($58.8M) and grew 72% YoY. The split between physical aftermarket parts and digital subscriptions (charging, theft tracking, battery analytics) is not disclosed. Moat scored the smart-platform proof quality at Low because the 4.54M connected-vehicle install-base figure is unaccompanied by monetization metrics. $11–21 of ARPU rising annually would upgrade the smart-platform moat from Low to Medium proof quality; negligible ARPU keeps it at install-base-as-marketing.
8. What is the FY2026 capex guide, and is the AIOS / Changzhou line on top of the $17–26M run-rate?
Status: unverified — needs external evidence. Capex/depreciation hit 1.53× in FY2025 ($25.4M vs $16.6M). Management has flagged AI-integrated product spend and a new factory line at Changzhou. Without explicit FY26 capex guidance, base-case FCF cannot be calculated. The $25M FY2025 FCF could compress to zero if capex steps another $11–14M higher; or it could double if depreciation absorbs the spend. This swing factor governs whether the 0.4× sales valuation is paying for FCF or for option value.
9. Is there a buyback authorization or dividend initiation pending?
Status: unverified — needs external evidence. At 0.4× sales with $154M of net cash ($189M cash less $35M debt), a buyback would be the highest-return capital deployment available to NIU. The board has not announced one. Inertia here is itself a signal — Quant ranked this as high priority because an announced buyback would be a material catalyst, and the absence of one despite multi-year share-price weakness raises a question about how management views fair value relative to $1.71 ADS grant prices.
10. Did the Best Buy US partnership deliver, or quietly wind down?
Status: unverified — needs external evidence. The Best Buy / US retail push dominated 2024 earnings calls but disappeared from the FY2025 narrative as international revenue fell 32.9% and Q4 international units fell 68.4%. Independent reporting on whether the Best Buy partnership generated material revenue or was reduced to a sub-scale presence would convert "the US retail strategy" from a strategic uncertainty into a closed chapter. Historian ranked this as high priority because the answer changes how investors weight management's strategic-pivot credibility.
Recent News Timeline
The pipeline did not return any external news. The table below records the dates of the only events visible from regulatory filings and company disclosures — events the specialists already had from the 20-F and proxy. There is no third-party news layer.
The five 13D/A amendments inside ten weeks of 2026 are the densest cluster of filings in NIU's history. The pattern itself — unaccompanied by any company press release — is the most actionable data point on this timeline.
What the Specialists Asked
Each tab below lists the queries a specialist filed to the web-research phase. Every one was rejected upstream before execution. The "Confidence" column is the same for every row — No evidence retrieved — because the page would mislead the reader to mark any of these otherwise. The questions themselves remain useful: they identify exactly where the filing-only thesis is most exposed.
Governance and People Signals
The web layer did not retrieve any third-party governance evidence. The table below summarizes what is visible from the filings themselves — the same data the Sherlock tab draws on. It is included here so the reader can see, in one place, the residual governance picture the web research would have been asked to deepen.
The pattern across grants. The three executive RSU events visible in the disclosures are all timed to depressed share prices. Without performance-vesting hurdles disclosed and without an external proxy-advisor critique to consult, the page cannot determine whether this is a deliberate alignment design or a one-sided pay structure. Sherlock ranked the proxy-advisor question as low priority because the answer is unlikely to move the thesis — but it remains an open verification item.
Industry Context
The web layer did not return external industry evidence. The Industry tab covers the primer; this section only flags the external industry questions that remain open and therefore where the filing-only industry view is most exposed. None of the items below could be triangulated against third-party data in this run.
What the page would do with a working web layer. A successful research run would (a) attach 13D/A amendment text to confirm whether Glory Achievement Fund is selling, (b) attach industry trade-press citations to either corroborate or refute Q3 2025 dealer pull-forward, (c) attach Honda EV-motorcycle volume data to size the largest external threat, and (d) attach analyst notes to frame the FY2026 consensus. Each of those four items is the difference between a thesis based on filings and a thesis cross-checked against the market.