Better Fintech Stock: SoFi Technologies vs. Upstart

SoFi has rapidly scaled revenue (adjusted net revenue +126% from Q3 2022 to Q3 2025) and is shifting to strong profitability—adjusted net income was $227M in 2024 with management expecting $455M in 2025 after an adjusted net loss of $54M in 2023—while launching crypto trading and Lightning-network cross-border transfers. Upstart, built on a decade of in‑house AI lending, reported Q3 2025 transaction volume and revenue growth of 128% and 71%, with auto originations and HELOCs up 357% and 324% YoY, and executives forecasting GAAP net income of $50M for 2025 versus a $129M net loss in 2024. Analysts favor Upstart on upside (consensus one‑year target implies ~24% upside) and a lower forward P/E (20.5) versus SoFi's forward P/E of 46.1, but the piece highlights higher execution and cyclicality risk at Upstart and a clearer profitability path at SoFi.

Analysis

Market structure: SoFi (SOFI) is consolidating as a profitable digital bank — winners include neo-banks, payment rails (Lightning integrations) and fintechs that scale deposits; losers are legacy scoring providers and branch-heavy banks losing retail deposits. Upstart (UPST) benefits lending partners via improved origination economics, but its partner-dependent model limits direct pricing power and concentrates execution risk. Cross-asset: stronger fintech profits should mildly tighten credit spreads (bps-level), lift risk assets and lower term premium in IG bonds; elevated idiosyncratic vols imply richer options premiums for both names.

Risk assessment: Key tail risks are a macro credit shock (NCO spike >1.5% leads to wide P/L swings), an AI/regulatory clampdown on algorithmic underwriting within 6–18 months, or operational/crypto losses at SoFi from Lightning/crypto exposure. Near-term (days–weeks) sensitivity centers on earnings prints and guidance; medium-term (3–12 months) on loan performance and funding costs; long-term (1–3 years) on cycle-tested profitability and regulatory regime for AI lending. Hidden dependency: Upstart’s revenue growth is levered to partner distribution and securitization capacity; SoFi is levered to deposit growth and cross-sell economics.

Trade implications: Core-risk: favor a quality fintech overweight (SOFI) as a lower-volatility core, and a capped-speculative tranche in UPST via options to express asymmetric upside. Use pair trades to isolate AI-lending adoption vs legacy scoring (long UPST, short FICO) and harvest volatility by selling covered calls on SOFI. Entry/exit signals: add on 8–12% pullbacks or positive sequential loan-performance beats; cut if 90+ delinquency rises >100 bps or guidance misses.

Contrarian angles: Consensus underestimates recession sensitivity of machine-learning models — Upstart could see outsized model drift under stress, making current valuation binary. Conversely, market may underprice SoFi’s cross-sell upside from crypto/payments monetization; a sustained RoTE >12% would justify >30% multiple expansion. Historical parallel: early online broker profit cycles — winners scale fast but incumbents regain share in downturns. Unintended consequence: broader AI-adoption invites stricter capital/reserve rules that compress ROE across fintechs.

Key Decisions for Investors

• Establish a 2–3% portfolio long position in SOFI (NASDAQ: SOFI) within the next 2 weeks or on a ≤10% pullback; time horizon 12–24 months, stop-loss at −20% and trim to half size if FY2025 adjusted net income guidance falls below $350M or 90+ delinquency increases >100 bps.
• Allocate 0.5–1% of portfolio to speculative UPST exposure via 6–9 month calls 30–50% OTM (size to limit downside); alternatively a 1–2% equity position sized small — add incrementally only after two consecutive quarters of positive loan-performance delta and GAAP profitability guidance (next 90 days).
• Implement a relative-value pair: long UPST (or calls) vs short FICO (ticker FICO) equal notional 1–12 month horizon (size 1% each) to play AI-adoption dispersion; unwind if Upstart misses credit metrics (90+ delinquency rise >100 bps) or FICO reports accelerating alternative-scorer wins.
• Hedge portfolio tail risk: buy 3–6 month SOFI puts equal to 5–10% notional of SOFI position or purchase protection on consumer-credit beta if Baa-10yr spread widens >50 bps; monitor regulatory proposals on AI credit over next 6 months and reduce leverage on any adverse rule announcements.