Novonix

Phase A — Understand the business

Lens 1 · Company Overview

Novonix Limited (ASX: NVX; Nasdaq: NVX via ADS, 1 ADS = 4 ordinary shares) is an Australian-incorporated, North-America-operated battery-materials company. Incorporated 2012 as Graphitecorp, ASX-listed 2015, renamed Novonix 2017 . HQ registered in Brisbane; management and 100% of operations sit in Chattanooga, Tennessee (anode) and Nova Scotia, Canada (battery tech). ~192 employees at 2025-12-31, 126 in the US .

What it actually is, as of mid-2026: a synthetic-graphite anode pure-play in the pre-commercial phase. The company reported two segments at year-end — NOVONIX Anode Materials (NAM) and Battery Technology Solutions (BTS) — but the BTS business was divested on April 30, 2026 to former CEO Chris Burns for $1, with Novonix retaining a 15% equity stake in the resulting cathode venture "Dryve Battery Materials" . It had already sold the Mt Dromedary natural-graphite exploration asset to Lithium Energy for A$2M in Sept 2025 . So the operating company is now essentially one thing: NAM's Riverside synthetic-graphite plant.

• Product / service: high-performance synthetic graphite anode material made via a proprietary continuous graphitization furnace process (vs the batch Acheson furnaces that dominate the Chinese incumbent base). Marketed on higher coulombic efficiency, capacity retention, purity ("essentially no contaminants"), and lower energy/emissions ``. Legacy BTS product = Ultra-High-Precision Coulometry (UHPC) battery-testing systems (Jeff Dahn's technology) — now divested.
• Customers: binding offtakes with Panasonic Energy (lead customer; mass production now guided to 2H 2027) and PowerCo SE (VW's cell arm; binding 32,000-tonne / 5-year offtake from 2027) . A joint-development agreement with LGES (which also holds a $30M convertible note). Sampling to ~13-15 prospective customers. Critically, NAM has generated $0 revenue to date ``.
• Suppliers: Phillips 66 is both a 10.57% shareholder AND the key supplier of petroleum needle coke (the graphite feedstock) . Harper International was the furnace-technology partner — but the exclusive license agreement **lapsed on Jan 1, 2026** because Novonix did not make the required initial payment (a red flag — see Lens 10/13).
• Contract structure: offtakes are market-price-formula, take-or-pay-style volume commitments but with product specs and qualification milestones still to be confirmed at future dates — i.e. not yet firm cash. The FCA/Stellantis offtake (86,250-115,000 tonnes) was terminated in Nov 2025 when the parties couldn't agree cell specs `` — proof these agreements can and do fall away pre-qualification.

Verdict on Lens 1: This is a technology-and-capacity call option on US anode onshoring, not an operating business. The revenue line ($5.6M FY25) came entirely from a business it has now sold.

Lens 2 · Supply Chain

Map the actual named chain, upstream → Novonix → end customer:

Upstream inputs

• Petroleum needle coke (primary anode feedstock) → Phillips 66 (also 10.57% holder; supplies coke and nominates a director) ``. Single strategic feedstock relationship = chokepoint #1.
• Graphitization furnace technology → Harper International Corporation — the exclusive-use/development relationship that underpinned the "proprietary continuous furnace" moat. The license expired Jan 1, 2026 for non-payment ``. Novonix says four furnaces are installed at Riverside, so hardware is in place, but the forward IP arrangement is now unresolved. Chokepoint #2 and a live risk.
• Electricity → Tennessee Valley Authority grid (>50% non-carbon: nuclear/hydro/wind/solar) — a genuine cost/ESG input advantage vs coal-heavy Chinese graphitization ``.
• Capital equipment / construction → general third-party EPC and equipment suppliers; $122.1M sat in construction-in-progress at 2025-12-31 ``.

Novonix (midstream conversion)

• Riverside facility, Chattanooga (404,000 sq ft, acquired Jul 2021) — scaling toward 20,000 tpa. Planned second facility "Enterprise South" (182-acre site, ~31,500 tpa initial) contingent on the DOE loan closing. Stated long-run ambition >50,000 tpa in Chattanooga, and a going-concern reference to "at least 150,000 tpa" ``.

Downstream customers / end market

• Cell makers: Panasonic Energy (→ Tesla/US autos), PowerCo SE (→ Volkswagen), LGES (JDA). End demand = EV batteries + grid ESS in North America.
• New diversification vector: industrial-grade synthetic graphite to "one of North America's largest value-add carbon processors" — shorter qualification, faster path to first revenue, targeted for 2026 ``.

Chokepoints / single-source dependencies: (1) Phillips 66 coke; (2) Harper furnace IP (now lapsed); (3) customer qualification — every offtake is contingent on passing each customer's unique cell-spec qualification, and Stellantis already walked. This chain is thin and fragile on both ends. Names present, so the lens clears — but the fragility is the story.

Lens 3 · Competitive Advantages (moats)

The bull moat narrative and an honest scoring of each:

• "Only qualified US-based supplier of battery-grade synthetic graphite" . This is the central claimed moat — regulatory/qualification scarcity. **Real but fragile and now less valuable:** (a) qualification is per-customer and revocable (Stellantis); (b) the *policy* leg that made "US-based" a durable premium — a punitive tariff wall on Chinese material — **was knocked out by the ITC in March 2026** (see Lens 8/10). The residual tariff protection is ~35% (Section 301 + a temporary global duty) , meaningful but not the 160-170% moat the thesis assumed.
• Process technology (continuous graphitization furnace). Claimed higher energy efficiency and yield vs batch Acheson furnaces. Genuinely differentiated if it scales at spec and cost — but the enabling Harper license lapsed Jan 2026 ``, muddying ownership of the very IP the moat rests on.
• IP estate: 8 issued patents + 14 active patent families at 2025-12-31; foundational cathode-synthesis patents granted in US/EU/China in 2025 ``. Note the strongest fresh patents are on the cathode tech that mostly went out the door with BTS/Dryve.
• Scientific bench: Dr. Jeff Dahn (Chief Scientific Advisor; also a Tesla research partner to 2026) and Dr. Mark Obrovac — top-tier lithium-ion researchers ``. Real reputational capital; harder to monetize now that testing/cathode (their nearest applications) is divested.
• Bargaining power: essentially none today. Novonix is a small, cash-hungry, pre-revenue supplier negotiating with giants (Panasonic, VW/PowerCo, Phillips 66). "Who needs whom more" points the wrong way — it needs the offtakes and the coke and the DOE money far more than any of them need Novonix. The market is explicitly described as "a small number of large dominating buyers" ``.

Moat verdict: A narrow, policy-dependent qualification moat, weakened materially by (i) the ITC ruling removing the punitive tariff, and (ii) the Harper license lapse clouding the process-IP claim. Not a wide moat. Optionality on a moat if the DOE loan closes and duties are somehow reinstated.

Lens 4 · Segments

Segment revenue is ``, and the punchline is stark — 100% of revenue comes from the segment being sold, and the go-forward segment earns nothing:

BTS FY25 split: Hardware $2.20M (point-in-time) + Consulting $3.42M (over-time) . BTS geography FY25: North America 78% / Asia 15% / Europe 7% .

Trend: total revenue has declined three years running ($8.1M → $5.9M → $5.6M), driven by softer BTS consulting demand ``. More importantly, post-divestiture the revenue line goes to ~$0 until NAM ships (guided 2H 2027), with a possible small bridge from industrial-grade graphite samples in 2026. Investors are buying a company that will report near-zero revenue for 18+ months while burning ~$40M+/yr. Segment lens = decisive: the accounting P&L is now almost irrelevant; the pre-production capacity plan is the company.

---

Phase B — Measure performance

Lens 5 · Earnings Result (FY2025, the latest full-year print)

All figures ``, USD, IFRS:

• vs consensus: n/a (thinly covered micro-cap; no reliable consensus line found ``).
• What drove the loss deeper: the $92.7M loss (vs $74.8M) is dominated by two non-cash/financing items — a $32.9M loss on extinguishment of the Yorkville convertible notes (fair value of shares issued > carrying value on conversion) and a $7.3M PP&E impairment, plus borrowing costs tripling to $11.9M ``. Underlying operating cash burn ($42.2M) is the number that matters and it is rising.
• Margins: meaningless at this stage — revenue barely covers direct product cost; the entire cost base is corporate + R&D + capacity build.
• Guidance/outlook: no financial guidance (pre-revenue). Operational guidance: Panasonic mass production 2H 2027; industrial-grade graphite 2026; expansion toward 20,000 tpa (Riverside) then 50,000+ tpa (with Enterprise South) ``. Tone: builder's optimism wrapped around a going-concern disclosure.
• Balance-sheet flags: cash $79.9M (up from $42.6M, entirely via financing — $95M Yorkville converts + $25M equity); net current assets only $5.6M (down from $11.1M); accumulated losses $352.4M; total borrowings carrying $93.8M ``. Receivables fell to $2.1M (from $8.2M) — consistent with a shrinking BTS book, not a red flag.
• Market reaction: stock ranged A$0.32 (Apr 2025) to A$1.02 (Oct 2025) over the year ; trades ~$0.4486 / ~$96.7M market cap as of Jul 5, 2026 — i.e. market cap barely exceeds cash, the classic pre-revenue "priced for dilution and doubt" setup.

Lens 6 · Earnings Calls / management focus (sentiment trend)

No transcripts on the shelf (transcripts=0); 20-F is a foreign private issuer so there is no US-style quarterly call cadence — Novonix reports via 6-K and ASX releases. Reading management's narrative arc across the FY25 filing and 2026 6-Ks :

• Consistent throughline: "only qualified US supplier," "onshoring critical minerals," "DOE-backed," "binding offtakes." The remuneration letter frames 2025 as milestone-driven (first industrial-grade sample; Panasonic equipment commissioned; $100M Yorkville raise; Commerce preliminary tariffs) ``.
• What shifted / what they stopped saying: the FY25 filing leaned heavily on the Commerce preliminary AD/CVD tariffs (93.5-102.7% AD, 11.5%+ CVD) as thesis validation. By the March 2026 6-K the tone pivots to "stays course … after ITC rejects new duties" `` — management dropped the tariff-wall drumbeat and re-anchored on offtakes + DOE. That is a genuine sentiment/emphasis shift forced by an adverse external event.
• New team = new language: a wholesale executive turnover in 2025 (see Lens 9) reset the narrative toward "operational excellence / capital-project execution" — consistent with a company moving from science to scale-up (and from a scientist-CEO to an auto/manufacturing operator).

Sentiment trend: guarded-constructive, and forced to become more defensive on the tariff leg through 2026. Label `` on the 2026 emphasis.

Lens 7 · Comps

Peer set = US/North-America-focused graphite-anode developers. All are pre-revenue or barely-revenue, so EV/Sales, EV/EBIT, P/E, dividend yield and 5-yr ROE are n/a across the board — the only honest comparison is market cap vs capacity/funding/time-to-first-sale. Multiples are `` with date or n/a; never fabricated.

Sources: (NVX, NMG); (SYR); `` (GPHOF).

Read: Novonix is the smallest-cap of the credible North-American anode developers and, on timeline, is behind Syrah (Vidalia targets first commercial AAM in 2H 2026, a full year before Novonix's 2H 2027). NMG carries ~5x Novonix's market cap on a longer-dated but more-integrated (mine-to-anode) plan. The comps do NOT support a "scarcity premium" for Novonix on relative valuation — if anything the market is discounting Novonix for its balance-sheet fragility and the lost tariff catalyst. The relevant multiple here is price-to-cash (~1.0x) and EV/nameplate-tpa — n/a for peers, so left blank rather than guessed.

Lens 8 · Stock-Price Catalysts (moves >5%, and what the tape reacts to)

for the FY25 range; for dated events:

• Oct 2025 — high A$1.02: driven by (management's own words) US-government support signals + fear of China restricting critical-mineral supply ``. The tape reacts hardest to policy/tariff news, not fundamentals.
• Feb 11-17, 2026 — Commerce FINAL AD/CVD: AD 93.5% maintained, CVD raised to 66.68%; combined cash deposits 160.32-169.58% on Chinese anode material ``. Bullish spike on the print.
• Mar 12, 2026 — USITC NEGATIVE injury determination (the reversal): ITC voted that Chinese AAM does not materially retard the establishment of a US industry → Commerce will NOT issue the AD/CVD orders; the 160-170% duties do not take effect . **NVX fell ~6% on the news** . This is the single most important catalyst of the last year and it went against the thesis. Residual tariff protection ≈ 35% (Section 301 + temporary global duty) ``.
• Apr 9, 2026 — Riverside $103M 48C tax-credit certification (positive; but note Enterprise South was *not* selected for 48C).
• Apr 30, 2026 — BTS divestiture completed (pure-play pivot) ``.
• Jun 11, 2026 — Panasonic mass-production C-sample delivered `` (incremental positive milestone).

Pattern: Novonix's stock is a policy-and-milestone instrument. It moves on tariff decisions, DOE/48C news, and qualification milestones — far more than on the (non-existent) earnings. The critical implication: with the punitive tariff catalyst now removed, the next dominant catalyst is binary — does the $754.8M DOE EDF loan actually close?

---

Phase C — Judge people & books

Lens 9 · Management

`` unless noted.

• Mike O'Kronley — CEO & Managing Director (joined May 2025; MD Dec 2025). 30+ yrs automotive, 15 in Li-ion/battery materials; prior roles at A123 Systems (a cautionary name — A123 went bankrupt in 2012), Metaldyne, Robert Bosch. Michigan engineering degrees. Track record: a genuine battery-materials operator brought in to move Novonix from science to manufacturing scale-up — the right archetype for this phase, but unproven at Novonix and A123 on his CV cuts both ways.
• Robert Long — CFO (Sept 2024; also interim CEO Jan-May 2025). CPA, ex-EY audit; founder/CEO of PE-backed Bridges Consumer Healthcare; Sanofi/Chattem finance leadership. Solid finance/audit pedigree for a going-concern company that lives or dies on raising capital.
• Chairman Ronald Edmonds (Chair from Jul 2025; board since Oct 2022; was interim "Executive Officer – Finance" Aug 2024-Jun 2025). Former Chief Accounting Officer / VP Controllers & Tax of Dow ($45B sales). Heavyweight controls/accounting credibility on the board — reassuring given the going-concern and convertible-accounting complexity.
• Dwayne Johnson — COO (Nov 2025): 30+ yrs manufacturing / capital-project leadership, managed >$500M capex, 10 greenfield plants. Exactly the capex-execution profile the Riverside/Enterprise South build needs.
• Kimberly Heimert — Chief Legal & Risk Officer (Sept 2025): ex-Deputy Chief Counsel of DOE's Loan Programs Office, ex-GC of OPIC/DFC. A pointed hire — she knows the DOE loan machinery from the inside, precisely the relationship Novonix most needs to close.
• Advisors / relationships: Dr. Jeff Dahn (Chief Scientific Advisor; Tesla partner to 2026); Andrew Liveris (strategic advisor, ex-Dow CEO, on boards of Aramco/IBM/Lucid) — his son Nick Liveris is a director and former CFO (disclosed family relationship / related party — see Lens 10). Phillips 66 nominates a director (Suresh Vaidyanathan).

(1) Track record: the go-forward team is <18 months in seat — essentially a brand-new management team hired for the scale-up phase. No Novonix delivery record yet; strong external pedigrees. (2) Tenure & skin in the game — a real weakness: all directors + senior management as a group own just 0.9% (7.2M ordinary shares) ``. Insider alignment is thin; incentives are via performance rights (see Lens 10). (3) Capital allocation: the honest record is a serial dilution + serial write-down history — KORE Power investment written to $0 (−$15.3M FY24), $32.9M convertible-extinguishment loss FY25, $7.3M PP&E impairment; funded almost entirely by issuing equity and convertibles (shares outstanding 487M→682M wtd-avg in two years, 840M now). ROE/ROIC deeply negative throughout. (4) Red flags: related-party web around the Liveris family and Phillips 66 (Lens 10); the Harper license lapse for non-payment is a capital-discipline eyebrow-raiser. (5) Archetype: transitioned from founder/scientist-led (Chris Burns, now gone) to professional-manager/operator-led — appropriate for the manufacturing phase, but it means the current stewards did not build the technology and must now execute someone else's plan under a going-concern clock.

Lens 10 · Forensic Red Flags

Acting as a forensic analyst. `` unless noted.

• GOING CONCERN — explicit, from both management and the auditor. Note 1: "these conditions give rise to a material uncertainty which may cast significant doubt (or raise substantial doubt as contemplated by PCAOB standards) over the Company's ability to continue as a going concern." The PwC audit opinion carries an explicit "Substantial Doubt about the Company's Ability to Continue as a Going Concern" section ``. This is the dominant forensic fact — not a footnote, an audit-opinion-level flag. Auditor: PricewaterhouseCoopers, Brisbane, serving since 2018. No auditor changes, no restatements, no material-weakness attestation required (emerging growth company / not subject to 404(b)).
• Cash flow vs earnings: net loss $92.7M vs operating cash outflow $42.2M — the gap is almost entirely the non-cash convertible-extinguishment loss ($32.9M), impairment ($7.3M), and borrowing-cost accruals. So the accounting loss overstates the cash bleed; the cash burn (~$42M) is the number to underwrite. This is a case where non-cash items make earnings look worse than cash — the opposite of the usual "SBC flatters non-GAAP" concern (and note SBC was a $1.1M credit in FY25 from forfeitures on executive departures).
• Financing-instrument complexity / dilution overhang (the real balance-sheet risk): embedded conversion features on the Yorkville and LGES convertibles are carried as derivative liabilities ($7.2M current) . **Yorkville (YA II PN) holds 35M debentures / $40M convertible notes; on full conversion at 2025-12-31 it would hold 142.3M shares (~22% of the company)** . LGES converts at A$1.60. With 17.4M performance rights + 14.0M options + convertibles outstanding ``, the fully-diluted overhang on top of 840M shares is very large — dilution is the central mechanism by which existing holders get hurt even if the plant succeeds.
• Impairment history: KORE Power equity investment written to $0 (−$15.3M FY24, maintained $0 FY25); $7.3M PP&E impairment FY25 — evidence that management's prior capital deployments have not held value.
• Revenue recognition: clean and simple — hardware at point-in-time, consulting over-time by labor hours ``. Immaterial contract liabilities. No rev-rec aggression (there's barely any revenue to be aggressive with).
• Related parties ``: (a) Phillips 66 — 10.57% holder, coke supplier, director-nominee whose $57.5K fees are paid to Phillips 66 not the individual; participated in the Jan 2025 $5M placement; (b) Andrew Liveris (strategic advisor, father of director Nick Liveris) granted 3.0M options at $0.60 vesting on anode-sales tonnage milestones; (c) Liveris Technologies Pty Ltd paid $180K for Nick Liveris's consulting. None are individually large in dollar terms, but the Liveris-family concentration (advisor + director + consulting entity + personal shareholding) is a governance texture worth flagging.

Regulatory findings (required sub-section):

• SEC Litigation Releases / AAERs: None. `` — EDGAR EFTS search (LR + AAER) for "Novonix" over 2021-07-06 → 2026-07-06 returned 0 SEC findings.
• 10-K/20-F Item — Legal Proceedings: "The Company is currently not a party to any material legal proceedings." ``.
• Non-SEC enforcement (FTC/DOJ/FDA/etc.): web search surfaced no enforcement actions, consent decrees, fines or penalties against Novonix . The only material government interactions are *supportive* (DOE grant/loan, 48C credit) — which themselves carry compliance obligations and clawback/debarment risk if milestones or representations fail .
• Conclusion: No material regulatory or legal findings — verified via SEC EDGAR EFTS (LR, AAER), web search, and 20-F Legal Proceedings as of 2026-07-06. The forensic risk here is not fraud or litigation — it is solvency (going concern) and dilution, both fully and honestly disclosed.

---

Phase D — Project & stress-test

Lens 11 · Forward Projection

This is a pre-revenue company; a bottom-up EPS model would be false precision. The honest projection is a cash-runway-to-catalyst + scenario frame. No forecast.ts logged (unattended --watchlist, and no committed base case — appropriately). All with arithmetic shown; inputs/`` as marked.

Runway (the number that actually matters):

• Cash at 2025-12-31: $79.9M ``.
• Operating cash burn FY25: $42.2M, and rising (FY24 $40.4M) . Add investing capex: PP&E payments were **$60.2M** in FY25 , partly grant-funded ($29.0M investing grants received).
• Near-term contractual: borrowings of ~$63.3M fall due in the 1-2yr bucket (largely Yorkville, expected to convert to equity not cash) ; capex commitments $15.1M .
• Runway estimate: on operating burn alone (~$42-45M/yr) and ignoring growth capex, $79.9M cash ≈ ~18-21 months → roughly through 2H 2026 into 2027 ``. But the plan requires hundreds of millions of growth capex to reach nameplate capacity, so substantial additional funding is required well before the plant generates cash — exactly the going-concern conclusion. Novonix will almost certainly raise again (equity/convertible) in the next 12 months absent the DOE loan closing.

EPS scenarios FY2026-FY2028 (all losses; illustrative, not a firm model):

• FY2026 base: revenue near $0 (BTS divested; only sporadic industrial-grade samples); loss/share roughly ($0.06)-($0.10) as the extinguishment/impairment one-offs don't repeat but burn continues and share count rises ``. n/a for any consensus cross-check.
• FY2027: first Panasonic revenue in 2H → still a loss; magnitude depends entirely on ramp speed and dilution.
• FY2028: the first plausible path to positive gross margin if Riverside runs at ~20,000 tpa on-spec and offtake pricing holds — but net profitability is not a FY28 base case.

The projection that matters is binary, not an EPS line: Does the $754.8M DOE EDF loan close, and does Novonix bridge its equity need without crushing dilution? If yes → Enterprise South funds and the capacity story is real. If no → serial dilution and a scaled-back single-plant company. I decline to fabricate a point EPS estimate the data can't support.

Lens 12 · Bull vs Bear

Adversarial, institutional.

Bull case. Novonix is the only qualified US-based supplier of battery-grade synthetic graphite into a North American EV/ESS supply chain that structurally must de-risk off China. It holds binding offtakes with Panasonic and PowerCo (VW) and a JDA with LGES; it is backed by an unusual weight of US-government support — a $100M DOE CMEI grant ($42.3M claimed), a $103M 48C tax credit certified for Riverside, and a conditional $754.8M DOE loan for Enterprise South. It has a scientific bench (Dahn) and a differentiated continuous-graphitization process, and it trades at 1.0x cash ($96.7M market cap on ~$80M cash) — so the market is ascribing almost nothing to the plant. If the DOE loan closes and Panasonic qualification completes in 2027, this is a multi-hundred-million-dollar revenue platform bought near cash value. Even without the punitive tariff, ~35% residual duties + IRA-driven "made-in-America" sourcing preferences give a domestic price umbrella. Deep, patient, policy-tailwind optionality.

Bear case (2-3 permanent-impairment risks).

1. Going-concern / dilution spiral. The auditor flags substantial doubt; the company burns >$40M/yr and needs hundreds of millions more to reach scale. Every raise at ~$0.45 with a Yorkville overhang that converts to ~22% is permanently dilutive; existing holders can be right on the technology and still lose most of their value to the cap table.
2. The tariff keystone is gone. The whole "US-based scarcity commands a premium" thesis leaned on a punitive wall against Chinese graphite. The ITC voted negative in March 2026 — no AD/CVD orders. Chinese material (at ~35% total tariff, still far cheaper) can compete for the very customers Novonix needs; and cell makers have every incentive to keep a cheaper qualified source.
3. Execution / customer-qualification risk. Novonix has never mass-produced battery-grade graphite on spec, at cost, at scale. Stellantis already terminated its offtake over cell-spec disagreement. The Harper furnace license lapsed. First revenue is 2H 2027 — and Syrah (Vidalia) targets first commercial AAM a full year earlier (2H 2026), so Novonix could be beaten to qualified US volume.

Pre-mortem (18 months out, thesis broke — what happened?): Most likely story: the DOE loan did not close (new administration's changed posture, 48C not granted for Enterprise South, or a condition unmet), forcing Novonix to fund Riverside completion through repeated dilutive raises; a Panasonic qualification slip pushed first revenue past 2027; and with no tariff wall, pricing came in below the offtake formula assumptions. The stock re-rated toward liquidation-of-cash value.

Are multiples too high? No — there are no earnings multiples; at ~1.0x cash the equity is arguably cheap on optionality. The risk isn't over-valuation, it's value transfer to the cap table + binary financing outcome.

Contrarian view (what the market may be missing, both ways): The bearish consensus post-ITC ("no tariff, no thesis") may under-weight that (a) IRA/domestic-content sourcing rules and OEM supply-security still create a real non-tariff pull for a qualified US supplier, and (b) a $103M 48C credit at Riverside is now certified, de-risking that plant's economics. Conversely, the bullish crowd is under-weighting that the DOE loan has not actually closed and the balance sheet is on a going-concern clock. The market is probably roughly right to price it as a cheap, high-variance option.

Lens 13 · Devil's Advocate (short-seller)

Dismantling the bull case.

• What structurally breaks the model: it has no revenue model yet — it's a promise contingent on (a) closing government financing, (b) passing customer qualification, and (c) producing at competitive cost. Any one failing breaks it. The most dangerous single point: the DOE loan is conditional and unclosed, and Enterprise South was not awarded the 48C credit it expected `` — a crack in the financing edifice.
• Revenue concentration: ~100% of future revenue is concentrated in two customers (Panasonic, PowerCo). One has a lead role, both can walk pre-qualification exactly as Stellantis did. That is customer concentration at its most fragile — concentration in contracts that aren't yet cash.
• Why the moat is weaker than bulls think: "only qualified US supplier" is a timing advantage, not a structural one. Syrah's Vidalia is targeting first commercial AAM in 2H 2026 — ahead of Novonix. NMG is building mine-to-anode. The ITC ruling means Chinese incumbents (Btr, Shanshan, etc.) keep cost leadership at only ~35% tariff. The "qualified" status is per-customer and revocable.
• Most dangerous competitor bulls underestimate: Syrah Resources — further along to qualified US AAM volume and vertically integrated back to its own Balama graphite mine (feedstock security Novonix lacks; Novonix depends on Phillips 66 coke).
• Worst capital-allocation moves / incentives: KORE Power to $0; a $32.9M value-destructive convertible extinguishment; letting the Harper license lapse for non-payment; insiders own just 0.9% while diluting outside holders through convertibles that hand Yorkville up to ~22%. Incentive alignment is weak.
• Assumptions that must hold for today's ~$0.45 price: that the company remains a going concern (raises capital as needed), that the DOE loan or equivalent funding materializes, and that Panasonic qualification completes. Strip out the DOE loan and you have a single-plant, serially-diluting micro-cap.
• If growth disappoints 20-30% (here: if first revenue slips to 2028+ or a second customer walks): the equity likely trades toward net cash (~$80M and falling), i.e. materially below the current market cap once further dilution is priced.
• Single scenario that permanently impairs the business: DOE loan collapses and a key offtake (Panasonic) terminates over qualification — Novonix becomes a distressed capital-raiser with a half-built plant and no anchor customer. Plausibility: moderate. Not the base case, but not tail-risk either — the going-concern flag exists precisely because this path is real.

Lens 14 · Management Questions (15, ordered by information value)

1. The DOE EDF loan is still "conditional." What specific conditions precedent remain, what is the realistic closing timeline, and — given Enterprise South was not awarded the 48C credit — has the loan's size, pricing, or viability changed? (This is the binary; everything else is secondary.)
2. Absent the DOE loan closing in the next 12 months, what is your funding bridge — size, instrument (equity vs convertible), and at what point does the going-concern doubt resolve?
3. Post-ITC (no AD/CVD wall, ~35% residual tariff), what is your delivered-cost position vs Chinese battery-grade synthetic graphite at the customer's gate — do your offtake price formulas still clear at a gross margin?
4. What are the exact remaining qualification milestones and dates for Panasonic before 2H 2027 mass production, and what is the contractual consequence if a milestone slips?
5. Stellantis terminated over cell-spec disagreement. What changed in how you contract and qualify so Panasonic or PowerCo can't walk the same way?
6. The Harper continuous-graphitization license lapsed Jan 1, 2026 for non-payment. Do you own or control the furnace IP you're running at Riverside outright, and is there any residual claim or re-licensing cost?
7. What is the fully-diluted share count at plausible conversion of the Yorkville and LGES notes, and how do you think about protecting existing holders from the ~22% Yorkville overhang?
8. Riverside's path to 20,000 tpa on-spec: what yield and cost-per-tonne are you achieving on qualification runs today vs the model, and where is the biggest variance?
9. Syrah's Vidalia targets first commercial US AAM in 2H 2026 — a year before you. How does being second-to-market affect your customer pipeline and pricing?
10. Petroleum needle-coke feedstock is single-sourced to Phillips 66, who is also a 10.6% holder. What happens to your cost and supply if that relationship changes, and are there qualified alternative coke sources?
11. What is the FY2026 and FY2027 operating cash-burn guide now that BTS is divested, and what's the minimum cash balance you'll run before raising?
12. You retained 15% of Dryve (the divested cathode venture). What is that stake worth to shareholders and what is the intended path — carry, monetize, or re-consolidate?
13. Insider ownership is ~0.9%. What is the plan to increase management skin-in-the-game beyond performance rights?
14. IRA / domestic-content demand pull: quantify the volume of your offtake book that is specifically driven by "made-in-America" sourcing rules vs pure cost/performance — how durable is that if EV subsidy policy shifts again?
15. What is your realistic base-case year for the first positive consolidated gross margin, and what has to be true to hit it?

---