Retail Team, March 2018 – June 2018


I worked with a team of engineers and a product manager from beginning to end—defining problems, scoping down the work, executing on design, working on improvements, and working the engineers on polishing the final product.

The 2 steps for AffirmGo

User & Business Problems

Core checkout is not optimized for quick, low AOV purchases

As Affirm's main product, core checkout is not optimized for purchases under $300. When the product was first developed, the goal was to target direct-to-consumer mattress and furniture companies, many of which start at $800+. Because of this, there are clear limitations to the product.

  • There are too many steps in the checkout flow, and the burden is too heavy on the user.
  • Other financial startups are starting to compete in the same space.
  • Core checkout is only optimized for and AOV of $300+.

Defining GOals & Measuring success

Increasing merchant sales by way of higher conversion rates with AffirmGo

After we scoped user and business problems, we put pen to paper on how we would define goals and measure success with this product.

The 3 main goals we focused on were:

  • Increase merchant sales by way of higher conversion rates with AffirmGo.
  • Increase initiation rate of more than 0.05%.
  • Build deeper relationships with merchants in the low AOV space—mainly fashion focused companies.

Scoping an MVP

Workflows that would allow us to test our hypotheses

I explored a few workflows based on the problem that we defined. We tried to pare it down to the most essential workflows for an MVP.

Exploration of different flows

  • As someone who is new to Affirm, I want to be able to trust that it is a safe and honest company, and not be bombarded with miscellaneous fees.
  • As someone who is familiar with and returning to Affirm, I want to be able to complete my purchase quickly, and have the same ease of repayment as core checkout.
  • As someone who has a lower FICO, I want to be able to afford my item and not be turned down from my purchase.


AffirmGo is not a loan

Through competitive research, we found that one of the advantages was that, because this is a 3 installment loan under $250, AffirmGo technically does not count as a loan. Because of this, we were able to remove a lot of the language around loans, terms selection, and legal disclosures.

Comparison of Affirm core checkout vs. AffirmGo

Removing language around loans and legal disclosures helped users feel less intimidated while deciding whether or not to use Affirm as a payment method.

This also helped simplify the flow, and removed a lot of the heavy lifting we put on users.


Rethinking the review screen

Explorations of review screen

Because we were able to remove a lot of the complexity around taking out a loan for AffirmGo, I had the chance to think about how to redesign the review screen. Some things I thought about included how to make everything easier to understand when optimizing for speed, as well as how to make it easy for the user to understand each payment amount, while being clear and upfront about everything.

I landed on creating an accordion-style timeline, with details being able to expand and collapse so as not to overwhelm the user with information, and only expose the information when user wants to see it.

We'd be using the same skeleton as Affirm Core Checkout, but needed to make sure that it was clear that people knew that they were using a 0 interest, 3 payment product, so they would not be choosing terms. I explored some visual treatments of the "review" screen for this cause.


Payment information

The target audience for AffirmGo is the younger generation. Generally people who are more interested in staying up-to-date in fashion and being trendy. Because this is our intended audience, and these people usually have not yet built up their credit, we lowered our FICO requirement in order to include this demographic.

In order to protect the business, and provide a better experience, we offer users with a lower FICO score the option to pay over time—we just require an initial first payment.

Explorations of first initial payment requirement

Because of this, it was important for me to have the payment timeline and payment instrument entry on the same screen, so the user would be able to make a clear connection between the two.

The payment instrument entry shows upon landing on the review page. The dropdown will show if it’s a returning user with a payment instrument already on file, making the repayment as easy as the purchase.

Learning after launch

More improvements to be made

We've seen a lot of exciting metrics since launching this, and are still continuously iterating on the product. We currently have about 300 merchants with an AffirmGo integration. We've seen 163,000 authorized loans to date amounting to $27,000. These numbers are still not nearly as high as we'd like them to be, but we do see a 2.8% increase in authorized loans week over week.

Through our data, we've also seen that most of our users are in the subprime, near prime, prime, and superprime categories. This is great not only because these users will go through a smoother application, but also because it increases chance of better repayment. This also means that although people are using AffirmGo, there is still a big category of people who are not using us, and we'd love to conduct some research, find out the reasons why they hesitate to use AffirmGo, make further improvements, and get them on board.

Reach out to me if you'd like to learn more.

Updated June 2019