Upstart Holdings, Inc. (UPST) Q3 2025 Earnings Call Transcript
Access earnings results, analyst expectations, report, slides, earnings call, and transcript.
Overview
The earnings call summary and Q&A indicate strong financial metrics and optimistic guidance. The company is focusing on AI enhancements and new product growth, which are expected to drive market share leadership. Despite some model conservatism impacting conversion rates, credit performance remains strong. The positive sentiment is bolstered by a robust shareholder return plan and strategic partnerships, suggesting a likely positive stock price movement over the next two weeks.
Key Financial Performance
Transaction Volume 80% year-on-year growth. Reasons: Increased consumer demand and more than 2 million applications submitted in Q3, up over 30% from Q2.
Revenue 71% year-on-year growth. Reasons: Driven by increased transaction volume and consumer demand.
GAAP Net Income Grew by a factor of 6 over the prior quarter. Reasons: Improved profitability and operational efficiency.
Auto, Home, and Small-Dollar Loan Originations 300% year-on-year growth. Reasons: Expansion in product offerings and improvements in software.
Auto Retail Transaction Volume Grew more than 70% sequentially. Reasons: Doubling of live lending rooftops and expansion to 4 new states.
Total Revenue $277 million, up 71% year-on-year and 8% sequentially. Reasons: Growth in transaction revenue and interest income.
Fee Revenue $259 million, up 54% year-on-year. Reasons: Steady growth in servicing revenue and increased transaction volume.
Net Interest Income $19 million, higher than expected. Reasons: Strong return performance on a temporarily elevated loan balance.
Loan Transactions 428,000, up 128% year-on-year and 15% sequentially. Reasons: Increased consumer demand and new borrowers.
Average Loan Size $6,670, 12% lower than the prior quarter. Reasons: Borrowers requesting lower amounts, model caution, and a mix shift towards smaller loans.
Contribution Margin 57%, down 1 percentage point from the prior quarter. Reasons: Lower conversion rates leading to higher acquisition and onboarding costs.
GAAP Operating Expenses $253 million, flat compared to Q2. Reasons: Variable expenses increased with transaction volume, but fixed expenses decreased by 7%.
Adjusted EBITDA $71 million, ahead of expectations. Reasons: Outperformance in net interest income and reduced fixed costs.
You have reached the limit. Sign up to access full content
Get started
Operating Highlights
Small-dollar loans, auto, and home products: These offerings accounted for almost 12% of originations and 22% of new borrowers in Q3. Transaction volume for these products grew around 300% year-on-year.
Auto retail business: Doubled the number of live lending rooftops in Q3 compared to the prior quarter. Transaction volume grew more than 70% sequentially. Expanded to 4 new states and made significant software improvements.
Auto secured personal loan: A hybrid product gaining traction.
Home business: Offers best rates to prime borrowers compared to other fintechs by as much as 300 basis points. Continued process and automation breakthroughs.
Expansion in auto retail: Expanded to 4 new states in Q3.
Bank and credit union partnerships: Added 7 new partners in Q3, reaching a new all-time high in monthly available funding.
Capital market partnerships: Signed 9 new partner agreements in Q3, totaling 17 for the year. Renewed one of the largest partners for the second time.
AI model precision: Models adjusted to macroeconomic signals, reducing approvals and increasing interest rates. Improved calibration methodology to cut month-to-month volatility by 50%.
Loan pricing latency: Reduced end-to-end latency by 30% through parallelization.
Customer acquisition: Developed a proprietary technique to target marketing spend, showing a 50% uplift in incremental originations from the same spend.
Home loan automation: Increased automatic home loan approvals from less than 1% in June to 20% in October.
AI-powered credit leadership: Positioned as a leader in AI-powered credit, with over $50 billion in AI-powered loans since inception.
Always-on credit vision: Aiming for 24/7 credit access with precise underwriting and best rates.
Third-party funding for new products: Secured 17 partner agreements in 2025 to support scaling of new products.
You have reached the limit. Sign up to access full content
Get started
Risk or Challenges
Transaction Volume: Transaction volume on the platform was less than anticipated due to risk models responding to macroeconomic signals by reducing approvals and increasing interest rates, leading to a lower conversion rate.
Macroeconomic Conditions: The Upstart Macro Index (UMI) ticked up modestly in July and August, causing the risk models to tighten credit approvals. This reflects the challenges of navigating economic volatility.
Model Conservatism: The AI models exhibited temporary conservatism in response to macroeconomic signals, leading to reduced transaction volumes and impacting revenue expectations.
Loan Balance: The company is carrying a larger-than-normal loan balance on its books, which ties up capital and creates dependency on closing deals to reduce R&D-related balances.
Funding Environment: Competitive funding environment has led to compressed spreads on third-party capital, which could impact profitability.
Repayment Speeds: Rising repayment speeds, while a positive long-term credit signal, limit interest income from current loans and require higher coupons to compensate.
New Product Scaling: Multiple new products are exiting R&D and entering the scale-up phase, requiring third-party capital arrangements to reduce reliance on balance sheet funding.
Economic Uncertainty: Decelerating personal consumption growth and a complex labor market create uncertainties in credit health and consumer behavior.
You have reached the limit. Sign up to access full content
Get started
Guidance & Outlook
Revenue Projections: For Q4 2025, total revenues are expected to be approximately $288 million, consisting of $262 million from fees and $26 million from net interest income. For the full year 2025, total revenues are projected to be approximately $1.035 billion, with $946 million from fees and $89 million from net interest income.
Contribution Margin: Expected to be approximately 53% for Q4 2025.
Net Income: GAAP net income for Q4 2025 is projected to be approximately $17 million, with adjusted net income of approximately $52 million. For the full year 2025, GAAP net income is expected to be approximately $50 million.
Adjusted EBITDA: Projected to be approximately $63 million for Q4 2025, with an adjusted EBITDA margin of approximately 22% for the full year 2025.
Loan Origination and Funding: Plans to reduce R&D-related balance sheet holdings and shift to third-party capital arrangements for new products, enabling further scaling in 2026. Multiple agreements are expected to be finalized by the end of 2025.
Market Trends and Economic Conditions: The company anticipates stable macroeconomic conditions with decelerating personal consumption growth, full employment, and easing monetary policies. These factors are expected to improve consumer financial health and lower investor return requirements in 2026.
Product Development and Growth: Continued focus on automation and AI advancements to improve credit performance and operational efficiency. Newer products like small-dollar loans, auto, and home loans are expected to scale further in 2026, supported by third-party funding agreements.
You have reached the limit. Sign up to access full content
Get started
Shareholder Return Plan
The selected topic was not discussed during the call.
You have reached the limit. Sign up to access full content
Get started
Key Q&A
Q:Can you comment on the strong application demand in Q3 and how it ties into the guidance, which was below expectations?
A:Applications grew about 30% quarter-on-quarter, driven by marketing programs and cross-selling. However, the model became more conservative due to macro factors, which impacted transaction volume. The application volume was the strongest in 3 years, but it did not translate to expected volume in Q3.
Q:Have recent high-profile bankruptcies and negative credit events in the auto space impacted your expansion plans or customer conversations?
A:No direct impact was observed. The company has rigorous processes to underwrite dealerships and mitigate risks. While headlines create caution in the market, there has been no major issue or wholesale change in the market.
Q:What are you seeing in the super prime segment, and how does it relate to the model tightness and competition?
A:The model reacted to macro signals, impacting the super prime segment. Subprime consumers had a relatively low UMI, while mid-700s segments had higher UMIs. The segment is competitive, and pricing impacts were observed.
Q:Can you provide insights into the quality of leads from improved marketing channels and the impact of AI on customer acquisition?
A:AI improvements in customer acquisition led to a larger increase in applications, but conversion rates were lower due to a conservative credit model earlier in the quarter. The quality of applications was impacted by targeting individuals less likely to be approved.
Q:Would an improvement in non-prime auto delinquencies significantly impact Upstart's auto originations?
A:The company has seen good credit performance in auto and believes its models are well-calibrated. Improvements in the market could create opportunities, and the auto business is expected to contribute significantly in 2026.
Q:What is implied in the Q4 guidance regarding application volume growth and model conservatism?
A:The model's conservatism from Q3 will impact Q4, even though UMI improvements are materializing. The company remains conservative and expects some lag in the model's response to improving conditions.
Q:Is refinancing credit card debt still the primary driver of personal loan demand?
A:Yes, refinancing credit card debt remains the dominant use case for personal loans, but they are also used for various purposes like home improvement or buying used cars.
Q:How are conversations with potential funding partners for R&D products progressing, and has there been any contraction in demand from private credit partners?
A:Conversations are progressing well, with good appetite from partners. There has been no contraction in demand, and the company has enough portfolio seasoning to demonstrate credit performance. However, diligence processes have lengthened due to broader market noise.
Q:What factors are driving the changes in conversion rates, and is there a target level for stabilization?
A:The primary factor is the model's conservatism, which impacted approvals and conversion rates. The company has worked to reduce variance caused by macro calibration, aiming for less volatility in future conversion rates.
Q:What is driving the observed increase in repayment speeds, and what are the credit implications?
A:Repayment speeds have increased broadly, possibly reflecting improving consumer health. While this reduces immediate interest income, it is generally a positive indicator for credit performance.
Q:What caused the changes in conversion rates, and how does the application flow compare to competitors?
A:Conversion rate changes were primarily due to the model's conservatism, impacting approvals. The company does not have specific data on how its loans compare to competitors like LendingClub or SoFi.
Q:What are the day 1 economics or take rates for the HELOC product compared to the corporate average?
A:HELOC take rates are expected to be healthy but more modest than personal loans, with loan sizes significantly larger.
Q:Are there any credit performance issues in the book despite UMI trends?
A:No, credit performance has been exceptional, and the UMI adjustments are part of the model's learning system.
Q:What elements of the model showed weakness, and how are they expected to improve?
A:The model's conservatism reduced approvals and conversion rates, responding to elevated risk signals. Improvements have been made to reduce noise and enhance responsiveness to real changes.
Q:How should we think about exit rates for Q4 and run rates for 2026?
A:The company is optimistic about Q4 and expects strong growth in 2026. The model's conservatism is seen as a feature, not a bug, and the company believes its approach will yield long-term benefits.
Q:Why does the UMI suggest different trends compared to broader market observations?
A:UMI trends show subprime borrowers in good shape, while mid-700s segments have elevated default rates. The data reflects a U-shaped trend, with strong performance at the low and high ends of the credit spectrum.
Q:What drove the lower-than-expected engineering and G&A expenses?
A:Lower expenses were due to fixed expense discipline and mechanical adjustments like reduced bonus payouts based on a lower business outlook.
Q:What are the use cases for HELOC loans?
A:HELOC loans are general-purpose, often used for home improvement, debt retirement, or other needs. They are substitutes for personal loans in some cases.
Q:Why does the model differ from market trends in terms of credit performance and volumes?
A:The model is designed to respond faster and more precisely to real-time credit signals, which may differ from traditional metrics used by competitors. This responsiveness can lead to different observations and decisions.
Q:Review of Unclear Management Responses
A:Management avoided directly addressing why their model's conservatism differed from broader market trends, particularly regarding credit performance and volumes. They also did not provide specific details on the populations or geographies driving elevated risk signals in the model.
You have reached the limit. Sign up to access full content
Get started
Earnings Word Cloud
The most frequently occurring keywords in this quarter's earning call
AI credit
AI platform
CEO Chairperson
Chase Co
Co Redburn
Co Research
Division JPMorgan
Division Needham
Division Piper
Founder President
Gu Co
JPMorgan Chase
LLC Research
Mizuho Securities
Needham LLC
Piper Sandler
President CEO
Redburn Research
Research Division
Rothschild Co
Sandler Co
Securities USA
Transaction volume
agreement partner
credit Upstart
demand
home loan
industry
latency
model calibration
partner agreement
signal
strength
technique
transaction volume
volatility
volume auto
UPST Transcript
Upstart Holdings, Inc. (UPST) Presents at Morgan Stanley US Financials Conference 2026 Transcript
Neutral6-10
Upstart Holdings, Inc. (UPST) Presents at J.P. Morgan 54th Annual Global Technology, Media and Communications Conference Transcript
Neutral5-19
Upstart Holdings, Inc. (UPST) Q1 2026 Earnings Call Transcript
Positive5-6
The earnings call presents a positive outlook with strong financial metrics like $294 million in adjusted EBITDA and positive net income. The company's strategy to reinvest in high-margin segments and focus on capital efficiency indicates a long-term growth trajectory. The launch of Cash Line and strong demand from banks and institutional investors further bolster this sentiment. Although there are higher expenses and some uncertainties, the company's robust funding strategy and positive market reception suggest a likely stock price increase in the short term.
Upstart Holdings, Inc. (UPST) Presents at Morgan Stanley Technology, Media & Telecom Conference 2026 Transcript
Neutral3-3
UPST Slides
Upstart Q4 2025 slides: Revenue hits $296M despite EPS miss, automation reaches 91%2026-02-10
Upstart Q3 2025 slides: Revenue soars 71% YOY, automation reaches 91%2025-11-04
Upstart Q2 2025 slides: first GAAP profitable quarter with 102% revenue growth2025-08-05
Upstart Q1 2025 slides: 89% origination growth, but stock tumbles on margin concerns2025-05-06
UPST Report
Frequently Asked Questions
Where does this earnings call transcript come from?
All transcripts are sourced directly from the official live webcast or the company’s official investor relations website. We use the exact words spoken during the call with no paraphrasing of the core discussion.
How soon is the transcript available after the earnings call ends?
Full verbatim transcripts are typically published within 4–12 hours after the call ends. Same-day availability is guaranteed for all S&P 500 and most mid-cap companies.
Is the transcript edited or altered in any way?
No material content is ever changed or summarized in the “Full Transcript” section. We only correct obvious spoken typos (e.g., “um”, “ah”, repeated 10 times”, or clear misspoken ticker symbols) and add speaker names/titles for readability. Every substantive sentence remains 100% as spoken.
Why do some answers appear as “Unclear” or “Inaudible”?
When audio quality is poor or multiple speakers talk over each other, we mark the section instead of guessing. This ensures complete accuracy rather than introducing potential errors.
Who creates the AI Summary and Key Q&A highlights shown above the transcript?
They are generated by a specialized financial-language model trained exclusively on 15+ years of earnings transcripts. The model extracts financial figures, guidance, and tone with 97%+ accuracy and is regularly validated against human analysts. The full raw transcript always remains available for verification.
