Affirm raises growth floor to 30% with new AI underwriting model
The buy now, pay later lender is using a transformer-based credit model to approve more shoppers without increasing its loss rate.
The short version
- Affirm raised its full-year growth floor for fiscal 2027 to at least 30%.
- Gross merchandise volume (GMV) grew 34% year over year in the first quarter of fiscal 2027.
- A new transformer-based credit model was trained on 14 years of transactional data.
- The new AI model delivered a 3.4% incremental conversion lift while keeping credit losses flat.
- Active merchants on the platform jumped 48% year over year to 621,000.
Why it matters
Affirm is proving it can use AI to expand its customer base without taking on more risk. This creates a competitive advantage by allowing the company to offer more approvals to merchants while maintaining a stable balance sheet.
Raising the growth floor
Affirm has officially raised its growth floor for fiscal 2027 to a minimum of 30%. This move comes on the heels of a strong first quarter where the company reported a 34% year-over-year increase in gross merchandise volume. Gross merchandise volume, or GMV, represents the total dollar value of all purchases processed through the platform.
Management expressed significant confidence in this trajectory after reviewing early trends from the second quarter. According to the company's results discussion, early data suggests that GMV growth may continue to stay in the high-30% range. This performance signals that the demand for the platform's services remains robust as the company scales.
The company also reported that revenue rose 35% during the first quarter, reaching between $1.26 billion and $1.3 billion. These figures provide a foundation for the updated outlook, suggesting that the company is successfully capturing a larger share of the shopping market. As reported by Crypto Briefing, the company is positioning itself for sustained expansion.
The transformer-based model
At the heart of this growth is a new AI underwriting model based on a transformer architecture. Transformers are a type of neural network architecture often used to process complex sequences of data. Affirm trained this specific model using 14 years of transactional data to better understand repayment behaviors and creditworthiness.
The new system is currently live in the United States and is designed to outperform the company's traditional underwriting methods. Affirm stated that the model is expected to improve approval rates by roughly 2 to 3 percentage points. Crucially, the company notes that these higher approval rates are being achieved without adding any additional risk to the balance sheet.
The model has already demonstrated its effectiveness by delivering a 3.4% incremental conversion lift. Conversion lift refers to the increase in the percentage of shoppers who complete a purchase after being approved for credit. By keeping credit losses flat while increasing approvals, Affirm is creating a more efficient way to manage its lending portfolio.
Merchant and market expansion
The growth of the platform is also being driven by a massive increase in merchant participation. Active merchants on the platform jumped 48% year over year, bringing the total to 621,000. This expansion allows Affirm to reach a wider variety of products and shopping categories for its users.
Despite this growth, Affirm currently holds a relatively small share of the total e-commerce market, which sits under 4%. The company currently operates in only four markets, leaving significant room for geographical and digital expansion. Management is looking to leverage this space to hit long-term growth targets that exceed the new 30% floor.
To capture more of this market, the company is moving beyond simple online checkouts. Affirm is currently pushing into software for offline merchants and exploring integrations with artificial intelligence. These initiatives aim to make the platform more versatile for businesses that do not operate exclusively in the online space.
Building a competitive moat
For investors, the ability to raise guidance after just one quarter is a meaningful signal of demand. It suggests that the company's recent performance was not a one-time lucky streak but rather a reflection of a durable trend. The high-30% growth seen in early Q2 adds further weight to the belief that the company can meet its new targets.
The underlying technology provides a significant competitive advantage that is difficult for rivals to replicate quickly. Because the credit engine is built on 14 years of specific repayment behavior, it serves as a data moat. Competitors cannot easily buy a similar historical dataset off the shelf to train their own models.
This technological edge also changes how Affirm interacts with its merchant partners. Instead of simply asking merchants to trust that consumers like installment plans, Affirm can now offer a measurable conversion lift. Providing concrete data on how many more shoppers will complete a purchase makes the service a much easier sell for retail businesses.
What to watch
- Whether the 2-3 percentage point approval lift holds steady as the model scales to more diverse products.
- How quickly Affirm can expand into more than the current four markets.
- The success of their new software offerings for offline merchants.
Try it
Look for the 'Affirm' logo on your next online purchase to see if their new AI model has expanded the types of products you qualify for.
Sources and links
Liked it? Say why
Add a line for other readers.
No account needed. We never show your email. Comments with links, ads or abuse are held and not shown. Your comment shows after a quick check.
Listen to the AI Wire headlines
Listen to the AI WireListen to our daily AI shows every morning: on your commute, in the shower, wherever you are
Listen to the daily showsFirst reported by Crypto Briefing · Read the original. Written by the SI Agent News Desk from the reporting above.Written with AI from the source above. Every number is checked against the source. Spot a mistake? Corrections.
